STRENGTH AND DUCTILITY OF AXIALLY LOADED RC SHORT COLUMNS

1

STRENGTH AND DUCTILITY OF AXIALLY LOADED RC SHORT

COLUMNS CONFINED WITH CFRP AND GFRP WRAPS

by

Haider Osamah Al-Karaghool

A Thesis Presented to the Faculty of the

American University of Sharjah

College of Engineering

in Partial Fulfillment

of the Requirements

for the Degree of

Master of Science in

Civil Engineering

Sharjah, United Arab Emirates

January 2013

2

© 2013 Haider Osamah Al-Karaghool. All rights reserved.

3

Approval Signatures

We, the undersigned, approve the Master’s Thesis of Haider Osamah Al-Karaghool

Thesis Title: Strength and Ductility of Axially Loaded RC Short Columns Confined with

CFRP and GFRP Wraps

Signature Date of Signature

___________________________ _______________

Dr. Adil K. Tamimi

Professor

Department of Civil Engineering

Thesis Advisor

___________________________ _______________

Dr. Jamal A. Abdalla

Professor


Thesis Co-Advisor

___________________________ _______________

Dr. Sherif A. Ahmed

Associate Professor


Thesis Committee Member

___________________________ _______________

Dr. Basil M. Darras

Assistant Professor

Department of Mechanical Engineering

Thesis Committee Member

___________________________ _______________

Dr. Sameh M. El -Sayegh

Head, Department of Civil Engineering

___________________________ _______________

Dr. Hany El-Kadi

Associate Dean, College of Engineering

___________________________ _______________

Dr. Yousef Al-Assaf

Dean, College of Engineering

___________________________ _______________

Dr. Khaled Assaleh

Director of Graduate Studies

4

Acknowledgment

Being a successful person is not that easy to be. To achieve something in our life,

challenges and difficulties have to be eliminated in order to achieve the greatest success in

our lives. As a student who spent four years of undergraduate studies and two years of

graduate studies in the American University of Sharjah, my life was full of challenges and

difficulties that without the help and support of so many individuals, I wouldn't reach the

person I am today.

First and foremost, I would like to thank my parents and family who supported me

throughout my educational life. In addition, special thanks to my advisor Dr. Adil Tamimi,

who acted as a senior design project advisor in undergraduate studies and a thesis advisor

in graduate studies, for his continues and enormous support and guidance over the past

years of study in the university.

In addition, I would like to thank Dr. Jamal Abdalla for being my co-advisor and

Dr. Sherif Ahmed and Dr. Bassil Darras for being my thesis defense committee. Also, I

would like to thank the faculty and staff of Civil Engineering department which includes

Dr. Sameh El Sayegh, Dr. Akmal Abdelfatah, Dr. Magdi El-Emam, Dr. Md. Maruf

Mortula, Eng. Arshi Faridi, Eng. Aqeel Ahmad, and Eng. Riyad Tamam. Also, I would

like to thank Ahmed A. Ghadban, Ahmad H. Al Rahmani, Mohannad Z. Naser, Adi S.

Abo-Obidah, Manal W. Kaakani, Rana A. Al Haje, Rami A. Al Haj, Noha T. Amer and

Assia Lasfer

.

5

Abstract

With the evolution of technology, construction industry has witnessed

enormous advancement in the production of concrete and construction materials. Such

materials have helped the community to produce structures that consume less energy,

environmental friendly, have less carbon foot print and are more durable with longer

life expectancy. However, old buildings that didn’t follow the latest technology are no

longer considered to have the previous qualities. In addition, they are no longer

considered safe due to the deterioration of the concrete, steel or both. As a result,

researchers continued to investigate different types of materials that can best

strengthen and rehabilitate aging concrete structures and render them safe. In this

investigation, RC short columns have been strengthened with CFRP and GFRP wraps

to increase their load carrying capacity and enhance their ductility. Thirty columns

were tested – fifteen of them were normally designed, i.e., they have the minimum

number of transverse reinforcement and the other fifteen were under designed, i.e.,

their transverse reinforcement is less than the minimum. The parameters investigated

were the type of material used for strengthening and the number of layers of CFRP

and GFRP used for wrapping the columns. Comparison of gains of axial strength and

ductility are presented in this research. It is observed that the axial load capacity of the

columns has increased, in the average, by 9.44 % and 24.56% when wrapped with one

and two CFRP layers, respectively while it has increased with, in the average, by

8.06% and 17.68% when wrapped with one and two layers of GFRP layers,

respectively. It is also observed that the ductility of the columns has increased when

wrapping with one or two layers of CFRP. It is also observed that the Ductility of the

columns has increased by 2.06% and 8.15% when wrapped with one and two layers of

CFRP, respectively while it has increased by 11.77% and 52.67% when wrapped with

one and two layers of GFRP layers, respectively. It can be concluded from this study

that CFRP wraps enhances the axial load capacity more than GFRP wraps while

GFRP wraps enhances the ductility more than the CFRP wraps.

Keywords: CFRP, GFRP, RC column, Strength, Ductility, Axial Load, Transverse

Reinforcement, Transverse Strain

6

Table of Contents

Abstract .......................................................................................................................... 5

List of Tables ............................................................................................................... 11

List of Figures .............................................................................................................. 12

CHAPTER 1 INTRODUCTION .............................................................................. 16

1.1 Problem Statement ........................................................................................ 16

1.2 Thesis Objectives .......................................................................................... 16

1.3 Scope of Work ............................................................................................... 18

1.4 Thesis Structure ............................................................................................. 18

CHAPTER 2 LITERATURE REVIEW ................................................................... 20

2.1 Introduction ................................................................................................... 20

2.2 Fiber Reinforced Polymers (FRP) ................................................................. 20

2.3 Properties ....................................................................................................... 21

2.4 Factors affecting FRP properties: .................................................................. 22

2.4.1 Effect of Moisture .................................................................................. 22

2.4.2 Effect of Alkalinity: ............................................................................... 22

2.4.3 Effect of Temperature ............................................................................ 22

2.4.4 Creep/Relaxation.................................................................................... 23

2.4.5 Fatigue.................................................................................................... 23

2.5 Advantages and Limitation of FRPs ............................................................. 24

2.5.1 Advantages of using FRP Composite Wraps ......................................... 24

2.5.2 Limitations of using FRP Composite Wraps ......................................... 24

2.6 Economic Considerations .............................................................................. 24

2.7 Confining RC Columns ................................................................................. 25

2.7.1 General ................................................................................................... 25

7

2.7.2 Methods of Confinement ....................................................................... 25

2.7.3 Failure Modes for Rectangular Columns ............................................... 27

2.7.4 Typical stress-strain curves of confined concrete .................................. 28

2.8 Ductility of FRP-Wrapped Columns ............................................................. 28

2.9 Previous Experimental Studies...................................................................... 29

CHAPTER 3 EXPERIMENTAL SETUP ................................................................ 36

3.1 Columns Configuration ................................................................................. 36

3.2 Main and Transverse Reinforcement ............................................................ 37

3.3 Concrete Mix Design .................................................................................... 38

3.4 Columns Preparation ..................................................................................... 39

3.5 Strain Gauge Fixing ...................................................................................... 40

3.6 CFRP and GFRP Properties .......................................................................... 41

3.7 Epoxy Preparation ......................................................................................... 41

3.8 Proposed Matrix ............................................................................................ 43

3.9 Columns Designation System ....................................................................... 43

3.10 Instrumentation and Testing Procedure......................................................... 44

CHAPTER 4 DISCUSSION OF EXPERIMENTAL RESULTS ............................ 45

4.1 N Group ......................................................................................................... 45

4.1.1 Load vs. Deflection ................................................................................ 45

4.1.2 Stress vs. Strain Diagram ....................................................................... 46

4.1.3 Transverse Strain ................................................................................... 46

4.1.4 Ductility Index ....................................................................................... 46

4.2 NC1 Group .................................................................................................... 47

4.2.1 Mode of Failure...................................................................................... 47

4.2.2 Load-Extension Diagram ....................................................................... 49

4.2.3 Stress-Strain Diagram ............................................................................ 50


8

4.2.5 Ductility Index ....................................................................................... 51

4.3 NC2 Group .................................................................................................... 51

4.3.1 Mode of Failure...................................................................................... 51




4.3.5 Ductility Index ....................................................................................... 55

4.4 NG1 Group .................................................................................................... 55

4.4.1 Modes of Failure .................................................................................... 55




4.4.5 Ductility Index ....................................................................................... 59

4.5 NG2 Group .................................................................................................... 59

4.5.1 Mode of Failure...................................................................................... 59




4.5.5 Ductility Index ....................................................................................... 63

4.6 U Group ......................................................................................................... 63




4.6.4 Ductility Index ....................................................................................... 65

4.7 UC1 Group .................................................................................................... 65

4.7.1 Mode of Failure...................................................................................... 65




4.7.5 Ductility Index ....................................................................................... 69

4.8 UC2 Group .................................................................................................... 69

9

4.8.1 Mode of Failure...................................................................................... 69




4.8.5 Ductility Index ....................................................................................... 73

4.9 UG1 Group .................................................................................................... 73

4.9.1 Mode of Failure...................................................................................... 73




4.9.5 Ductility Index ....................................................................................... 77

4.10 UG2 Group .................................................................................................... 77

4.10.1 Mode of Failure...................................................................................... 77




CHAPTER 5 THEORETICAL RESULTS .............................................................. 82

5.1 Introduction to ACI 440.2R .......................................................................... 82

5.2 Design Equations........................................................................................... 82

5.3 ACI Prediction............................................................................................... 85

5.3.1 CFRP Confinement ................................................................................ 85

5.3.2 GFRP Confinement ................................................................................ 86

5.4 Experimental vs. Theoretical Results ............................................................ 87

CHAPTER 6 ANALYTICAL MODEL ................................................................... 89

6.1 Existing Models of FRP-Confined Concrete ................................................ 89

6.2 Prediction of a New Model ........................................................................... 89

6.2.1 Normal Design with CFRP .................................................................... 91

6.2.2 Normal Design with GFRP .................................................................... 91

6.2.3 Under Design with CFRP ...................................................................... 92

10

6.2.4 Under Design with GFRP ...................................................................... 92

6.3 Verification of the New Models .................................................................... 93

CHAPTER 7 SUMMARY OF RESULTS ............................................................... 94

7.1 Summary of the Lab Work ........................................................................... 94

7.2 Summary of the Mode of Failure .................................................................. 94

7.3 Summary of Load and Ductility Results ....................................................... 95

7.3.1 N Group ................................................................................................. 95

7.3.2 U Group ................................................................................................. 96

7.3.3 N vs. U Group ........................................................................................ 97

7.4 Summary of Transverse Strain results .......................................................... 97

7.4.1 N Group ................................................................................................. 97

7.4.2 U Group ................................................................................................. 98

7.4.3 N vs. U Group ........................................................................................ 98

Conclusion ................................................................................................................... 99

Proposed Future Work ................................................................................................. 99

References .................................................................................................................. 100

VITA .......................................................................................................................... 104

Appendix A Extra Figures and Diagrams ............................................................ 105

Appendix B Notations ......................................................................................... 117

11

List of Tables

Table 2-1: Qualitative Comparison of FRP materials [6] ............................................ 22

Table 2-2: A comparison of different methods of column strengthening .................... 27

Table 3-1: Concrete Mix Design Proportions .............................................................. 38

Table 3-2: Average load and strength results of concrete cubes and cylinders ........... 39

Table 3-3: Testing Results of FRP materials provided by the manufacturer ............... 41

Table 3-4: Properties of Primer.................................................................................... 41

Table 3-5: Properties of Saturant ................................................................................. 42

Table 3-6: Matrix designation system.......................................................................... 44

Table 5-1: Excel Sheet for the calculations of the confined load ................................ 85

Table 5-2: Theoretical Calculations-CFRP confined column ...................................... 86

Table 5-3: Theoretical Calculations-GFRP confined column ..................................... 87

Table 5-4: Theoretical vs. Experimental Results (Percentage Difference For N-Group)

...................................................................................................................................... 88

Table 5-5: Theoretical vs. Experimental Results (Percentage Increase for N-Group) 88

Table 6-1: Analysis of the experimental results .......................................................... 90

Table 6-2: Predicted values comparisosns ................................................................... 93

Table 7-1: Summary of Mode of Failure ..................................................................... 95

Table 7-2: N Group Load and Ductility Results .......................................................... 96

Table 7-3: U Group Load and Ductility Results .......................................................... 96

Table 7-4: N vs. U Group Load and Ductility Comparisons ....................................... 97

Table 7-5: N group Transverse strain comparisons ..................................................... 97

Table 7-6: U group Transverse strain comparisons ..................................................... 98

Table 7-7: N vs. U group Transverse strain comparisons ............................................ 98

12

List of Figures

Figure 2-1: AFRP [2] ................................................................................................... 21

Figure 2-2: CFRP [3] ................................................................................................... 21

Figure 2-3: GFRP [4] ................................................................................................... 21

Figure 2-4: Full Wrap .................................................................................................. 26

Figure 2-5: Partial Wrapping using discrete rings ....................................................... 26

Figure 2-6: Partial wrapping using continues spiral .................................................... 26

Figure 2-7: Typical Stress-Stain Curves of FRP confined square columns [24] ......... 27

Figure 2-8: Typical Stress-Strain Curves of FRP confined Concrete [28] .................. 28

Figure 3-1: Normal Design Reinforcement ................................................................. 37

Figure 3-2: Under Design Reinforcement .................................................................... 37

Figure 3-3: Normal design detailing for the reinforcement ......................................... 37

Figure 3-4: Under design detailing for the reinforcement ........................................... 37

Figure 3-5: Rebar Sample Stress vs. Strain curve........................................................ 38

Figure 3-6: Strength of Concrete Cubes ...................................................................... 39

Figure 3-7: Strength of Concrete Cylinders ................................................................. 39

Figure 3-8: Top view-original shape............................................................................ 39

Figure 3-9: Top view-rounded shape ........................................................................... 39

Figure 3-10: Column Configuration ............................................................................ 40

Figure 3-11: Strain Gauge Fixing ................................................................................ 40

Figure 3-12: Primer Part A .......................................................................................... 42

Figure 3-13: Primer Part B ........................................................................................... 42

Figure 3-14: Primer Effect [24] ................................................................................... 42

Figure 3-15: Saturant part A ........................................................................................ 43

Figure 3-16: Saturant Part B ........................................................................................ 43

Figure 3-17: Testing Equipment .................................................................................. 44

Figure 3-18: Specimen under testing ........................................................................... 44

Figure 4-1: N Load vs. Extension Diagram ................................................................. 46

Figure 4-2: N Axial vs. Transverse Strain ................................................................... 47

Figure 4-3: NC11 Two Sides view .............................................................................. 48

Figure 4-4: NC11 One Side view................................................................................. 48

Figure 4-5 : NC12 corner view ..................................................................................... 48

13

Figure 4-6: NC12 One side view ................................................................................. 48

Figure 4-7: NC13 Corner View ................................................................................... 49

Figure 4-8: NC13 Corner View ................................................................................... 49

Figure 4-9: NC1 Load vs. Extension Diagram ............................................................ 50

Figure 4-10: NC1 Axial vs. Transverse Strain............................................................. 51

Figure 4-11: NC21 Corner View ................................................................................. 52

Figure 4-12: NC21 Side View ..................................................................................... 52





Figure 4-17: NC2 Load vs. Extension Diagram .......................................................... 54

Figure 4-18: NC2 Axial vs. Transverse Strain............................................................. 55

Figure 4-19: NG11 Corner View ................................................................................. 56

Figure 4-20: NG11 Side View ..................................................................................... 56




Figure 4-24: NG13 Closer Side View .......................................................................... 57

Figure 4-25: NG1 Load vs. Extension Diagram .......................................................... 58

Figure 4-26: NG1 Axial vs. Transverse Strain ............................................................ 59


Figure 4-28: NG21 Top view ....................................................................................... 60




Figure 4-32: NG23 Different Side View ..................................................................... 61

Figure 4-33: NG2 Load vs. Extension ......................................................................... 62

Figure 4-34: NG2 Axial vs. Transverse Strain ............................................................ 63

Figure 4-35: U Load vs. Extension Diagram ............................................................... 64

Figure 4-36: U Axial vs. Transverse Strain ................................................................. 65

Figure 4-37: UC11 Corner View ................................................................................. 66

14

Figure 4-38: UC11 Close Corner View ....................................................................... 66




Figure 4-42: UC13 Close Corner View ....................................................................... 67

Figure 4-43: UC1 Load vs. Extension ......................................................................... 68

Figure 4-44: UC1 Axial vs. Transverse Strain............................................................. 69

Figure 4-45: UC21 Side View ..................................................................................... 70

Figure 4-46: UC21 Close Side View ........................................................................... 70


Figure 4-48: UC22 Top View ...................................................................................... 70


Figure 4-50: UC23 Side View ..................................................................................... 71

Figure 4-51: UC2 Load vs. Extension ......................................................................... 72

Figure 4-52: UC2 Axial vs. Transverse Strain............................................................. 73

Figure 4-53: UG11 Corner View ................................................................................. 74

Figure 4-54: UG11 Different Corner View ................................................................. 74


Figure 4-56: UG12 Side View ..................................................................................... 74



Figure 4-59: UG1 Load vs. Extension ......................................................................... 76

Figure 4-60: UG1 Axial vs. Transverse Strain ............................................................ 77







Figure 4-67: UG2 Load vs. Extension ......................................................................... 80

Figure 4-68: UG2 Axial vs. Transverse StrainDuctility Index .................................... 81

Figure 5-1: Equivalent circular section ........................................................................ 84

15

Figure 6-1: NC Analytical Model ................................................................................ 91

Figure 6-2: NG Analytical Model ................................................................................ 91

Figure 6-3: UC Analytical Model ................................................................................ 92

Figure 6-4: UG Analytical Model ................................................................................ 92

Figure A-1: N vs NC1 ................................................................................................ 105

Figure A-2: N vs NC2 ................................................................................................ 105

Figure A-3: N vs NG1................................................................................................ 106

Figure A-4: N vs NG2................................................................................................ 106

Figure A-5: U vs UC1 ................................................................................................ 107

Figure A-6: U vs UC2 ................................................................................................ 107

Figure A-7: U vs UG1................................................................................................ 108

Figure A-8: U vs UG2................................................................................................ 108

Figure A-9: NC1 vs UC1 ........................................................................................... 109

Figure A-10: NC2 vs UC2 ......................................................................................... 109

Figure A-11: NG1 vs UG1 ......................................................................................... 110

Figure A-12: NG2 vs UG2 ......................................................................................... 110

Figure A-13: NC1 vs NG1 ......................................................................................... 111

Figure A-14: NC2 vs NG2 ......................................................................................... 111

Figure A-15: UC1 vs UG1 ......................................................................................... 112

Figure A-16: UC2 vs UG2 ......................................................................................... 112

Figure A-17: N vs UC1 .............................................................................................. 113

Figure A-18: N vs UC2 .............................................................................................. 113

Figure A-19: N vs UG1.............................................................................................. 114

Figure A-20: N vs UG2.............................................................................................. 114

Figure A-21: NC1 vs NC2 ......................................................................................... 115

Figure A-22: NG1 vs NG2 ......................................................................................... 115

Figure A-23: UC1 vs UC2 ......................................................................................... 116

Figure A-24: UG1 vs UG2 ......................................................................................... 116

16

CHAPTER 1 INTRODUCTION

1.1 Problem Statement

Gulf Capital Countries (GCC) are well known for their hot climate and high-

humidity levels, going as high as 60°C and 80% during summer. Due to the high

temperature ranges, water evaporates causing the humidity level to rise. Upon

evaporation, the water consists of salts called “airborne salt” which adversely affect

the strength property of concrete by penetrating the surface. This effect takes place

when, upon penetration, they induce a chemical decomposition that results in

corroded reinforcements that degrades the strength of the structure. This drop in

concrete strength is a serious cause of concern. Consequently, the average life-cycle

of a concrete structure is considerably shorter when compared to the surrounding

regions.

Besides corrosion, other factors such as changes in the use of a structure and

new design codes also cause structure deficiency. Changes in structures occur upon a

change in live loads, for example an increase in traffic load due to traffic congestion.

New design codes, on the other hand, can potentially classify some buildings as

deficient.

Due to the difficulties associated with the production of high strength and

impermeable concrete mixtures that resist corrosion of steel rebars, a critical need for

innovative and well-engineered solutions has arisen. Fiber Reinforced Polymers

(FRP) is the solution put forth, with applications that range from the maintenance and

rehabilitation of deteriorating RC structures to the construction of new projects that

were once perceived as architecturally-challenging. High strength, high durability,

high corrosion resistance, high strength-to-weight ratio, ease of site installation,

electrochemical neutrality, and fire resistance are some of many encouraging pros that

make FRP materials the most favorable choice for strengthening structures.

1.2 Thesis Objectives

The origin of this idea was the need to find an alternative solution that would

not consider demolishing structures classified as unsafe. The outcome of this research

provided a better understanding of the FRP wrapping systems when applied on

17

structural concrete elements. This study examined the improvements of these systems

on the compressive strength and ductility. The primary objectives behind this research

were:

1. Studying the behavior of Normal-design RC column when strengthened with

CFRP and GFRP in one or two layers.

2. Studying the behavior of Under-designed RC columns when strengthened with

CFRP and GFRP in one or two layers.

3. Comparing the behavior of strengthened under-designed RC columns with the

non-strengthened Normal-design RC column.

4. Predicting the maximum confined concrete compressive strength based on

ACI-440 and compare it with the experimental results for Normal-design

columns.

5. Developing a parametric study that predicts the compressive strength of

confined concrete based on the properties of the wrapping system.

This study investigated the behavior of RC column in terms of strength and

ductility. In addition, it provided an indication as to whether or not the strengthening

of the under-designed RC column can be considered the alternative solution to

demolishing columns in the concrete building.

18

1.3 Scope of Work

This thesis evaluated strengthening reinforced concrete short columns with

CFRP and GFRP systems. This evaluation was based on the number of wrapping

layers and the type of design method followed for all the columns. Some of the

columns were designed to follow the minimum requirement based on ACI 318 code,

hence categorized as safe while the rest were designed based on reducing the number

of ties in order to ensure less strength handling. These columns were subjected to

different strengthening scenarios based on the wrapping material and number of

layers wrapped around the columns. After finalizing the wrapping process, columns

were tested to measure their load capacity and ductility.

1.4 Thesis Structure

This document is divided into seven chapters. Chapter 1 serves as the

introduction, providing a clear idea about the topic and a brief description of the

problem faced in the structural buildings. Lastly, it lists the objectives and the scope

of work covered in this research.

Chapter 2 serves as a literature review covering all the published research

materials dealing with the behavior of reinforced concrete members strengthened with

FRP systems. It starts with an introduction about FRP materials and their various

types. It also includes mechanical properties, advantages and limitations, and

economic considerations about the FRP materials. Finally, it summaries earlier

experiments carried out on structural concrete members strengthened with FRP

wrapping materials.

Chapter 3 discusses the experimental setup designed for this research. It

introduces the steel configuration of the two types of columns (Normal and Under)

according to ACI-318 code with a description of all the materials used (steel,

concrete, epoxy, CFRP and GFRP). This chapter also looks over the columns’ surface

preparation for strengthening with FRP wraps. Finally, it covers the proposed matrix

and the identification system for this research that was designed based on the

questioned parameters.

19

Chapter 4 examines the findings of the testing results for the entire matrix. For

each group, there is an overview and discussion about the mode of failure, the load vs.

deflection curve, the stress vs. strain curve, and the ductility index. The purpose of

these curves was to study the behavior of each column after testing.

Chapter 5 mentions the calculation process of the theoretical values for the

confined columns. The theoretical analysis was based on ACI-440.2R code. After an

introduction about the code, it describes the procedure followed in order to calculate

the theoretical confined concrete compressive strength. Furthermore, a spreadsheet is

introduced to demonstrate the inputs and outputs of the analysis. Sample calculations

to be compared to the actual results are also noted in this spreadsheet.

Chapter 6 provides the development of the analytical model, covering the first

model for predicting the theoretical value of compressive strength, and then recent

more accurate models created over the past few years. It also mentions the prediction

process of the new model that was used for all the cases that were considered. Finally,

a verification step was created to find how accurate the results of the new model were

when compared to the previous models.

Chapter 7 includes the summary for the entire body of work in this research,

illustrating the discussions for the mode of failure and the comparisons for the load,

deflection, and ductility results between all the groups.

20

CHAPTER 2 LITERATURE REVIEW

2.1 Introduction

In the last two decades, there has been an extensive demand for alternative

strengthening systems that would benefit structures in terms of strength and life-cycle

expectancy. Today, FRP composites are the most reliable material used for

strengthening structures, gaining this reputation due to various distinguished qualities

such as ease of installation, corrosion resistance, etc. While research is still ongoing

on such materials, not enough research is being carried out to study their strength and

ductility enhancement capabilities when applied on RC short columns (considered as

under-design columns). This chapter serves as a literature review for some of the

recent studies related to the behavior of different RC structural members subjected to

axial load.

2.2 Fiber Reinforced Polymers (FRP)

Fiber Reinforced polymer (FRP) is defined as composites resulting from

fusing two materials. One of the constituents is fibers, which are long strips of

fiberglass, aramid, or carbon. The other material is the polymer matrix. This material

serves as a binder that holds the fibers together to form the fiber reinforced polymer.

These binders can be found as epoxy, vinylester or polyester thermosetting plastic.

FRPs provide additional strength and stiffness to the structural members. Also,

they provide axial strength in the longitudinal direction and shear strength in the

transverse direction [1]. The following types of FRPs are most commonly used due to

their low cost, high thermal insulation, high tensile strength, and high fatigue

resistance, as opposed to steel or aluminum:

1. Glass (GFRP)

2. Carbon (CFRP)

3. Aramid (AFRP).

21

Figure ‎2-1: AFRP [2]

Figure ‎2-2: CFRP [3]

Figure ‎2-3: GFRP [4]

Fiber Reinforced polymers (FRP) are found in different shapes and lengths for

various engineering applications such as bars, plates and sheets. FRP bars. For

instance, it can replace steel rebars to serve as longitudinal or transverse

reinforcement in different structural members. FRP Plates and sheets, similarly, can

be used to retrofit, rehabilitate, or strengthen RC deteriorated or degraded structures.

2.3 Properties

The physical and mechanical properties of the matrix are the base for defining

the material behavior and characteristics of the FRP composite. Factors such as fiber

volume, type of fiber, type of resin, fiber orientation, dimensional effects, and quality

control during manufacturing play a big role in establishing the characteristics of an

FRP material [5].

22

Table ‎2-1: Qualitative Comparison of FRP materials [6]

Criterion Aramid Carbon Glass

Young Modulus Good Very Good Adequate

Tensile Strength Very Good Very Good Very Good

Compressive Strength Inadequate Very Good Good

Long-term Behavior Good Excellent Very Good

Stiffness Good Very Good Adequate

Fatigue Behavior Good Excellent Adequate

Bulk Density Excellent Good Adequate

Alkaline Resistance Good Very Good Inadequate

Price Adequate Adequate Very Good

2.4 Factors affecting FRP properties:

2.4.1 Effect of Moisture

The infiltration of moisture in the FRP composites adversely affects

performance [6, 7]. Water penetration into FRPs is divided into two phases:

1. Mixed with the resin

2. Penetration into the cracks.

The former occurs during the mixing of the epoxy, where water molecules

evaporate in the air due to the humidity level and mix with the epoxy, resulting in a

decrease in the quality of the resin. In case of the latter, the penetration of water or

any other flaws happen occurs due to capillary flow [8, 9].

2.4.2 Effect of Alkalinity:

Durability is crucial in the design of concrete structures. Concrete is

considered to be high in alkalinity (pH=12.8) potentially leading to a reaction between

fibers (mainly glass), resulting in a reduction in composite strength, stiffness and

strength [1].

2.4.3 Effect of Temperature

Changes in temperature directly affect the rate of moisture absorption and the

mechanical properties of FRP [10, 11, 12, 13]. With an increase in temperature, the

23

mechanical properties of FRP composites decrease while accelerating the creep and

stress relaxation. This can be very clear when the temperature reaches glass transition

(Tg – 30°F and above) [14].

A decrease in the temperature, on the other hand, does not cause a severe

decrease in mechanical properties [15]; however, it can lead to possible increases in:

1. Tensile and flexural strength.

2. Fatigue strength and creep resistance.

3. Modulus of Elasticity.

Additionally, a decrease in FRP temperature can lead to possible decrease in:

1. Elongation and deflection.

2. Fracture toughness and impact strength.

3. Compressive strength.

4. Coefficient of linear expansion.

2.4.4 Creep/Relaxation

Generally, the increase of creep strains occurs due to poor matrix properties

and curing percentage. Moreover, resins (polymer) viscoelasticity play a vital role in

affecting creep stains of FRP materials [1]. Carbon based FRPs do not creep like other

type of FRPs; however, GFRPs exhibit a poor behavior under sustained loading. As a

result, the tensile strength of GFRPs plummets (as low as 20% of maximum) when

the material is subjected to permanent tension.

2.4.5 Fatigue

FRP composites accumulate damage micro-structurally as the number of load

cycles increases. Micro-structural damage includes fiber/matrix debonding and matrix

micro cracking. The fatigue behavior of composites materials depends on the fabric

lay-up sequence, temperature, moisture content, frequency, and maximum to

minimum stress/strain ratio [1]. It is interesting to notice that CFRP exhibit superior

fatigue performance to Steel. In fact, the dominant factor in the fatigue of FRP-

strengthened members is the fatigue of existing steel reinforcement.

24

2.5 Advantages and Limitation of FRPs

2.5.1 Advantages of using FRP Composite Wraps

Higher strength-to-weight ratio (15 and 35, respectively, for glass and carbon,

compared to that of steel)

Higher stiffness-to-weight ratio (1 and 3, respectively, for glass and carbon,

compared to that of steel)

Higher corrosion resistance

Lighter unit weight, resulting in less-expensive equipment for economical

handling, shipping, and transportation as well as lighter erection equipment

Higher durability, leading to lower life-cycle costs

Greater ductility, providing ample warning before collapse

Easier-to-reinforce micro-crack zones

Easier-to-control tension crack growth by the confining concrete

Better customization for specific needs

Faster field installation, resulting in more economical procedures for the

confinement of concrete in columns than steel jacketing

Simpler field corrections in case of installation defects of bonding of FRP with

concrete substrate

2.5.2 Limitations of using FRP Composite Wraps

Uncertainties about the durability of FRPs, as data about their long-term

performance is limited

Concerns of fire resistance

Limited knowledge of material properties and application procedures

Possible continuation of steel bar corrosion in warped concrete members

Lack of adequate laboratory and field data with respect to various structural

actions, including the shear-lag phenomenon due to an increase in the number

of fiber composite wrap layers

2.6 Economic Considerations

Compared to steel and aluminum, CFRP plates and sheets are considerately

more expensive. The raw material cost, alone, is often four times that of steel.

25

However, installation costs, transport and handling costs are much lower as opposed

to steel installation costs. Most importantly, the installation of CFRP is quick. A

reduced contract program obviously lowers the ancillary costs of access and plant

hire, propping and sit set-up. Even more significant are the reduced timescales for

road closures or traffic management.

Thus, with negligible planned expenditure on maintenance, the economics of

CFRP has become very attractive with the high cost of the carbon fiber composites

being counterbalanced sufficiently.

2.7 Confining RC Columns

2.7.1 General

Strengthening RC columns is the most common applications of FRP as it

enhances the load carrying capacity, ductility, and transverse strain. In addition, the

lateral confinement increases the axial strength and ductility. On the other hand, the

change in the transverse strain decreases with the increase in the lateral confinement.

Until the 1990s, there were two methods used for confining RC columns [16]:

1. Reinforced concrete cage

2. Grout-injected steel jackets.

Steel jacketing is more effective than caging because it provides an increase in

the cross-sectional area and weight of the structure. However, both methods require

intensive work and are difficult to install. Furthermore, both caging and steel

jacketing are made of steel which means that they are highly vulnerable to corrosion

due to the low resistance against weather attacks [16, 17].

2.7.2 Methods of Confinement

Three types of confinement are illustrated in this: wrapping, filament winding,

and prefabricated shell jacketing.

2.7.2.1 Wrapping

FRP wrapping is considered to be the most common method of strengthening

RC columns with FRP composites. This method, known as the wet lay-up method,

involves unidirectional FRP wraps being fully submerged in epoxy. The direction of

26

the wraps is perpendicular to the axis of the column. The methods of wrapping are

different from each other. An RC column can be fully wrapped with FRP composite

in one or more layers as shown in Figure 2-4. It can also be either partially wrapped

using discrete ring of one or multiple layers as shown in Figure 2-5. It can even be

wrapped using continues spirals of one or more layers as in Figure 2-6.

Figure ‎2-4: Full Wrap

Figure ‎2-5: Partial Wrapping using

discrete rings

Figure ‎2-6: Partial wrapping using

continues spiral

The first demonstration for enhancing the compressive strength of confined

RC members with external FRP wraps was made by Fradis and Khalili [18, 19].

2.7.2.2 Filament Winding

The filament winding follows the same principle of wrapping, except it uses

continues fiber straps instead of discrete sheets allowing it to be processed

automatically via computer software. Through filament winding, an FRP jacket with

specific thickness, fiber orientation and volume fraction can be obtained. Fradis and

Khalili [18] were the first to introduce the confinement of concrete by winding

continues resin-impregnated fiber strands. The first winding machine was developed

in Japan in the mid-1980s [20].

2.7.2.3 Prefabricated Shell Jacketing

The shells are fabricated under controlled conditions using fiber sheets or

strands with the impregnation of resins affected prior to field installation. They can be

fabricated in half-circles, half-rectangles [21, 22] and circles with a slit or in

continuous rolls [23], so they can be opened and placed around columns. For effective

FRP confinement, a full contact between the column and the FRP shell is essential.

This is ensured by bonding the shell to the column using adhesives or by injecting

27

shrinkage-compensated cement grout/mortar into the space between the shell and the

column [21, 22].

Table ‎2-2: A comparison of different methods of column strengthening

Method Advantages Disadvantages

Wrapping

Flexibility in coping with

different columns shapes

Ease in site handling, without the

need for special equipment

Least quality control

Most labor intensive

Filament Winding Improved quality control

Reduced on-site labor

Reduced flexibility in

coping with different

columns shapes

Special equipment

required

Prefabricated

Shells

Best Quality control

Least on-site labor

Useful for column shape

modification

Limited flexibility in

coping with different

columns shapes

Prefabrication Cost

2.7.3 Failure Modes for Rectangular Columns

It is well-established that FRP confinement is less effective for rectangular

columns than for circular columns, despite rounding off the corners. The reasons for

this are that confining pressure is uniformly distributed and that only a part of the

concrete core is effectively confined.

Figure ‎2-7: Typical Stress-Stain Curves of FRP confined square columns [24]

Failure generally occurs at the corners by FRP tensile rupture. The

stress-strain curves are more likely to feature a descending branch and FRP

confinements provide little strength enhancement. In such cases, the ultimate strength

of the FRP-confined concrete is reached before the ultimate strain of FRP, where the

28

FRP-confined concrete carries a substantial stress which may or may not exceed the

strength of the unconfined concrete [25, 26, 27].

2.7.4 Typical stress-strain curves of confined concrete

The stress of unconfined concrete column increases as the load increases. A

decrease in the load occurs after the yield level is reached due to the compressive

failure of the concrete column. However, RC concrete column wrapped with FRP

behaves differently. When there is sufficient level of confinement, the axial load

increases due to the resistance of the FRP warps. This increase continues until it

reaches the ultimate compressive strength (point D shown in Figure 2-8). This is

known as “Strain Hardening”. If the level of confinement is insufficient or weak,

however; the columns can resist some of the applied axial load until it reaches a

dropping point (point C in the Figure 2-8). This case is called “Strain Softening”.

Figure ‎2-8: Typical Stress-Strain Curves of FRP confined Concrete [28]

2.8 Ductility of FRP-Wrapped Columns

Other than strength, ductility is considered to be equally important as strength

when studying the behavior of FRP-wrapped columns as they are affected by the

displacements at yield and failure point of the concrete column. Ductility can be

defined as a solid material's ability to deform under tensile stress. Hue et. al [29]

discussed the strength and ductility of partially deteriorated strength concrete columns

confined with CFRP. The calculations of the ductility for the confined concrete

columns are based on the deformation readings of the yield and failure points on the

load vs. deflection curve. Ductility (μ∆) is calculated based on the following equation:

29

Where represents the deflection at ultimate point and represents the deflection

at yield load in the load vs. deflection curve.

2.9 Previous Experimental Studies

In previous decades, extensive research was carried out on strengthening

reinforced concrete columns with various FRP materials in order to improve strength

and ductility. Such studies have helped develop standards for the design of concrete

mix with a specific compressive strength and type. Furthermore, the last century has

seen many methods being introduced for retrofitting reinforced concrete structures.

Such methods started off with using steel for such purposes and eventually migrating

to materials such as aluminum and FRPs.

R. Kumutha, R. Vaidyanathan, and M.S. Palanichamy [30] studied the

behavior of axially loaded rectangular columns strengthened with glass fiber

reinforced polymer (GFRP) wraps. The objectives of this study centered upon

evaluating the effectiveness of external GFRP strengthening for rectangular concrete

columns, evaluating the effect of number of GFRP layers on the ultimate load and

ductility of confined concrete, and evaluating the effect of the aspect ratio of the

column on the effectively confined cross-section. A total of nine specimens were

subjected to axial compression, including three control specimens. The specimens

were loaded to failure in axial compression and the specimen behaviour in axial and

transverse directions was investigated. The parameters of this study included the

aspect ratio of the cross-section (1, 1.25, and 1.66) and the number of GFRP layers (0,

1, and 2).

All nine reinforced concrete columns were also tested under concentric

loading and had the same dimensions: a length of 750mm and a cross-sectional area

of 15625mm2. The classification of columns followed certain designation represented

by three terms. The first term refers to the number of GFRP sheets making up the

jacket. The second term describes the shape of the column cross-section. ‘S’ refers to

a square cross-section and ‘R’ refers to a rectangular cross-section. The third term

which is a number in subscript refers to the aspect ratio of the column cross-section.

30

In conclusion, effective confinement with GFRP composite sheets resulted in

higher compressive strength. Better confinement was achieved when the number of

GFRP wrap layers was increased, resulting in enhanced load carrying capacity of the

column, in addition to overall improvement in ductility. The load carrying capacity of

the column decreased with an increase in aspect ratio of the cross-section. The test

results show a definite overall linear relationship between the strength of confined

concrete and lateral confining pressure provided by FRP.

Muhammad N.S. Hadi [31] presented the results procured by testing wrapped

columns subjected to eccentric loads. This paper provided a description of the loading

mechanism and results of testing nine prismatic circular columns tested under

eccentric load. The columns were wrapped with CFRP or GFRP. Nine short

cylindrical high strength concrete columns were designed for testing. Three columns

were reinforced with steel bars and the remaining six columns were made of plain

concrete. Three of the six plain columns were wrapped with unidirectional carbon

while the remaining three columns were wrapped with weave E-glass. The general

properties and the dimensions of column specimen are shown in the following table:

In his conclusion, external confinement with FRP composite appeared to

significantly augment the strength of concrete column. However, when the eccentric

load was introduced into the experiment, the strength loss was vastly evident. In

addition, the maximum load capacity of a confined column under eccentric load was

directly related to the magnitude of eccentricity. That is, a larger eccentricity results in

a smaller maximum load. However, the lateral deflection—another important design

criterion—had no direct relation with the eccentricities. Furthermore, externally

confined concrete column could undergo large deformation without rupture (the

extent of deformation could be decided by the strength of FRP composite). Finally,

when tested both concentrically and eccentrically, the CFRP wrapped columns

resulted in higher loads and ductility as opposed to GFRP-wrapped and steel-

reinforced columns

Omar Chaallal, Mohsen Shahawy, and Hunzer Hassan [32] presented the

results of a comprehensive experimental investigation on the behaviour of axially

loaded short rectangular columns strengthened with carbon fiber-reinforced polymer

(CFRP) wrap. The objectives of the study were assessing the effectiveness of external

31

CFRP strengthening for rectangular short concrete columns, assessing the effect of the

number of CFRP layers on the ultimate strength and ductility of the confined

concrete, assessing the effect of the aspect ratio of the column on the effectively

confined cross-section, and to monitor the influence of the compressive strength of

unconfined concrete on the gain in strength and ductility of the confined concrete. The

parameters considered in this study were:

1. The concrete strength (3 ksi and 6 ksi)

2. The aspect ratio of the cross-section (a/b= 1, 0,654, and 0.5)

3. The number of CFRP layers (0, 1, 2, 3, and 4).

As conclusion, the confinement provided by the CFRP improved both the

load-carrying capacity and the column ductility. This method of structural

rehabilitation was shown to be applicable to rectangular sections as well. In addition,

as the compressive strength of concrete increased, both the axial and transverse strains

decreased significantly. Square columns generally exhibited higher dilation ratios than

the rectangular columns. In fact, after a certain degree of confinement, the dilation

ratio decreases with respect to an increase in jacket stiffness. Furthermore, the

stiffness of the applied CFRP jacket is the key parameter in the external jacket retrofit

designing. The jacket must be sufficiently stiff to develop appropriate confining

forces at relatively low column axial strain levels. A stiff jacket also better controls

the dilation of the cross-section, resulting in larger axial strain capacities. Finally, a

gain in the compressive strength of CFRP confined concrete is governed by the

stiffness ratio of the FRP jacket between lateral direction and axial stiffness of the

column.

Manuel A.G. Silva [33] presented the results of the tests performed on axially

loaded RC columns (both square and circular cross-section) with and without jackets.

The FRP tested were made of either CFRP or AFRP (aramidic wraps). Moreover, a

comparison of gains of axial strength and ductility was presented along with aspects

of variation of the lateral pressure and FRP jackets ruptures. Tests were performed on

reinforced concrete cylinders and square prisms of 0.75m height and an aspect ratio

(height/diameter or width) equal to five. The prismatic columns of square cross-

section were divided into three groups according to corner sharpness:

1. R1 – sharp-edged corner

32

2. R2 – corner radius equal to 20mm

3. R3 – corner radius equal to 38mm, which corresponds to 1/4 of the width of

the square-section.

Columns with small chamfer, R0, were also tested.

In his conclusion, the improvement of axial load capacity from jackets of

AFRP or CFRP was equivalent for cylindrical columns. The improvement in ductility

could not be conclusively shown as higher for AFRP than for CFRP as they were

roughly similar for the cylindrical columns tested in this program. Columns of square-

section and sharp corners evidenced no improvement of capacity or ductility upon

confinement in CFRP jackets. For AFRP confinement, there was improvement of load

capacity, but no significant improvement in ductility. Lastly, estimated jacket rupture

lateral strains were considerably lower than ultimate strains obtained from flat

coupons owing to “strain localization phenomena” in the jacket. This is a result of

concentrated actions due to concrete crushing, buckling of longitudinal reinforcement

and rupture of stirrups.

Y. Toutanjii and Y. Deng presented an extensive research on axially loaded

members confined with AFRP [34]. They investigated the performance of AFRP

confined concrete columns in wet/dry and freeze/thaw conditions. A total of 24

cylindrical specimens were utilized, with dimension of 76mm x 305mm. 12 columns

were confined with AFRP and the other 12 were plain (control). Their mix design had

a water/cement ratio of 0.5, sand/cement ratio of 2.0, and gravel/cement ratio of 3.0.

The course aggregate contained crushed stone with a maximum size of 12.7mm with

fine aggregate composition of 50% river and 50% beach sand. All the specimens were

prepared via 28 days of curing at 25oC and 90% humidity. The average compressive

strength of concrete, for the 28-days curing, was 44MPa. The concrete cylinders were

wrapped with two layers of unidirectional AFRP composites and all samples were

loaded at a loading rate of 0.24MPa/s in uniaxial compression until failure.

The experimentation led to the conclusion that AFRP confinement constrains

the lateral strain producing a tri-axial stress field in the concrete. This improved the

compressive strength, maximum strain, and ductility of the columns. In addition,

durability test results indicated that wet/dry environment had little effect on the

compressive strength of AFRP-wrapped specimens. And finally, exposure to

33

freeze/thaw environments had marginal effect on the compression strength of AFRP-

wrapped columns.

Hua Wei, Zimin Wu, Xia Guo, and Fumin Yi [35] conducted an experimental

study on partially deteriorated strength concrete columns confined with CFRP, aiming

to study the mechanical behavior of deteriorated parts confined with CFRP. They

proposed two series of columns:

P-Series consisted of 15 columns in five groups with plain concrete in total.

Each group had three identical specimens. Each specimen was 150mm x

150mm in cross-section and 550mm in height. Each specimen also had three

segments with two grades of concrete strength: lower-strength casted at the

middle while the higher-grade casted at the top and bottom of the column.

S-Series consisted of five reinforced concrete columns. Each specimen had

200mm x 200mm x 1250 mm with 4 main reinforcement bars of 14mm

diameter, yield strength of 335MPa, and ultimate strength of 555MPa. They

also consisted of stirrups of 6mm diameter with yield strength of 345MPa,

lateral spacing of 150mm at the test part and 50mm lateral spacing at the

edges. Similar to P-Series columns, four columns were divided into three

segments with two grades of concrete strength and the fifth column had higher

grade of strength.

The letter “P” indicated columns with plain concrete and “S” indicated

reinforced column. “U”, similarly, marked unwrapped columns while “W” marked

wrapped column. Unwrapped column with a deteriorated strength parts (middle part)

were identified by “U1” while “U2” identified unwrapped columns with a single

higher compressive strength.

After testing, they concluded that partial confinement in deteriorated regions

with CFRP significantly enhances the performance of columns in terms of strength

and ductility. The load capacity of the entire column can also subsequently be

improved.

The ductility of confined specimens was enhanced significantly compared to

the partial deteriorated columns and the original columns. The gain in load capacity

34

was different with the layer of CFRP. The greater the number of CFRP layers, the

greater the gain in load capacities. Partial confinement on deteriorated parts can be

developed as an alternative approach in axial compressive conditions to avoid cost

and time-consumption.

Raafat El-Hach, Mark F. Green, and Gordon R. Wight [36] studied the Effect

of Severe Environmental Exposures on CFRP-wrapped concrete columns. In their

paper, the aim was to study the behavior of CFRP wrapped concrete cylinders when

subjected to harsh environmental conditions (heating-cooling cycles, freezing-thawing

cycles, and exposure to fresh and salt water) and compare this to the experimental

results with predicted values of ACI 440 and ISIS Canada 2001. The experimental

program involved testing 36 standard plain concrete cylinders (150mm diameter by

300mm in length) in six different environments closely simulating aggressive

conditions. In each group, three cylinders were confined by wrapping them with

epoxy bonded CFRP sheets at room temperature and three were left unwrapped. The

cylinders were wrapped twenty-two days after casting. All cylinders were tested to

failure in axial-compression at room temperature. The environmental conditions were:

1. Room temperature +20±3°C and relative humidity 50%±5% for 70 days.

2. Heating and cooling cycling (+23 to +45°C) for 33 cycles. Each cycle had a

maximum temperature of 45°C that was maintained for about 24 hour then

decreased to room temperature for another 24 hours.

3. Prolonged exposure to high temperature +45±5°C for 70 days. In this group, 3

cylinders were wrapped with the CFRP sheets before they were subjected to

the high temperature exposure. The other three were wrapped after exposure to

high temperature.

4. Heating-cooling cycling (+23 to +45°C) for 22 cycles. Each cycle had a

maximum temperature of 45°C maintained for roughly 24 hours then

decreased to room temperature for another 24 hours. This was followed by 33

freezing and thawing cycles. The freezing-thawing was performed by placing

the concrete cylinders in the cold room overnight at −18°C for 16 hours, and

removing them in the morning to thaw in a water bath at +18°C for 8 hours.

35

5. Heating-cooling cycling (+23 to +45°C) for 22 cycles followed by immersion

in fresh water (pH=8 at 23°C) for 33 days.

6. Heating-cooling (+23 to +45°C) for 22 cycles followed by immersion in salt

water pH=9 at 23°C for 33 days. The sodium chloride (NaCl) was a 3.5% by

weight solution to simulate exposure to seawater.

They concluded that with a confining wrap of two layers, the strength of the

concrete cylinders increases by up to 43 and 74% over unconfined cylinders kept at

room temperature and subjected to heating-cooling cycles. Axial strain in the confined

cylinders was approximately four times greater. Furthermore, heating-cooling cycles

reduced the compressive strength for the unwrapped cylinders. No significant

difference in strength between the wrapped cylinders subjected to heating-cooling and

the specimens kept at room temperature was found. Freezing-thawing exposure as

well as fresh-salt water immersion also had a slightly negative effect on the

compressive strength of both unwrapped and wrapped cylinders as opposed to room

temperature and heating-cooling exposure. Finally, the predicted ultimate FRP-

confined concrete compressive strength using the ACI 440 and ISIS Canada models

compared favorably with the experimental results.

Although many researchers have investigated the effect of FRP materials on

enhancing the performance of RC structures in normal and severe environmental

conditions or in circular and rectangular shapes, topics concerned with enhancing RC

columns with lower transverse reinforcement has not met researchers’ interest. In this

research, the behavior of these columns was investigated when enhanced with CFRP

and GFRP.

36

CHAPTER 3 EXPERIMENTAL SETUP

3.1 Columns Configuration

Two reinforcement configurations were used in this research. The first

configuration considered the normal design which followed the minimum

requirements set by ACI 318 when designing an RC short column. This group had 15

columns which shared the same characteristics. Each column had a standard cross-

section of 150mm x 150mm and a length of 750mm. Based on the minimum

requirements of the ACI 318, 4 rebars of No. 10 were used for the main reinforcement

of the column and No. 10 spaced at 140mm were used for the transverse

reinforcements.

The other group had the same configuration except for the spacing of the

transverse reinforcement, that is, each stirrup is spaced at 234mm instead of 140mm.

The reason behind increasing spacing and decreasing the number of stirrups is to

study the level of enhancement contributed by the FRP materials. This amount of

spacing was based on the assumption of 50% reduction of the minimum number of

reinforcement. The aim is to study the decrease in load when the number of transverse

is reduced to half the minimum, and compare this with the normal-design columns

which met the minimum requirements.

In order to predict the theoretical load of the column, an analysis was created

to measure how much load can the unconfined concrete column undergoes when axial

load is applied. Using ACI 318 code, the following equation was used to calculate the

theoretical load:

[ ( ) ] -----------------------------------------------Equation 1

In Equation 1, the concrete compressive was found to be 45MPa at 28 days

curing, the cross-sectional area was calculated to be 22500 mm2, the area of steel was

314.16mm2 for 4 #10 mm rebars, and the yield strength of the steel was measured to

be 600N/mm2. All these parameters were used to analyze the column and calculate the

load at which the column fails. The load was found to be 830kN.

37

Figure ‎3-1: Normal Design Reinforcement

Figure ‎3-2: Under Design Reinforcement

Figure ‎3-3: Normal design detailing for the

reinforcement

Figure ‎3-4: Under design detailing for the

reinforcement

3.2 Main and Transverse Reinforcement

ASTM E8 code was followed in testing the performance of steel rebars. A No.

10 steel rebar sample was tested to measure its yield strength and ultimate strength.

The total length of the specimen was 300mm. The sample was tested under a

deformation rate of 10mm/min. From Figure 3-5, the steel rebar reached yield strength

of 600N/mm2. In addition, the elasticity modulus was found to be 200kN/mm

2.

4 # 10 4 # 10

#10 @ 140

mm

#10 @ 234

mm

38

Figure ‎3-5: Rebar Sample Stress vs. Strain curve

3.3 Concrete Mix Design

The concrete was made of four primary constituents: cement, water, coarse

aggregate, and fine aggregate. The concrete mix was designed with strength of

45MPa. Ordinary Portland cement (Type 1) was used with a specific gravity of 3.14

for the concrete production. The coarse aggregate was divided into 65% of sieve size

passing 20mm and 35% of sieve size passing 10mm. In addition, fine aggregate was

divided into 60% of crushed sand and 40% of dune sand. The composition of the

concrete mix was designed by weight and described in the following ratios.

Table ‎3-1: Concrete Mix Design Proportions

Material Cement Water Coarse Aggregate Fine Aggregate

Ratios by weight 1 0.4 2.92 1.65

S.G 3.14 1 2.61 2.57

The columns were casted horizontally in ply-wood forms. Later, the concrete

was evenly distributed using a vibrator in order to decrease the air volume in the

concrete when poured in the forms. Following this, all samples were leveled with a

spatula to ensure that the side surface of the columns were as smooth as possible. The

final stage of the casting involved covering all the columns with a plastic sheet to

avoid any water loss during the curing.

39

Three 150mm x 150mm cubes and three 100mm x 200mm cylinders were

tested after 7, 14, and 28 days of curing to measure the strength of the concrete after

casting,

Table ‎3-2: Average load and strength results of concrete cubes and cylinders

Cube Cylinder

Days Load (KN) Strength (Mpa) Load (KN) Strength (Mpa)

7 630 28.0 171 21.8

14 830 36.9 223 28.4

28 975 43.3 263 33.5

Figure ‎3-6: Strength of Concrete Cubes

Figure ‎3-7: Strength of Concrete Cylinders

3.4 Columns Preparation

All columns had a round edge with a radius equivalent to 25mm in the axial

direction as shown in Figure 3-8 and Figure 3-9. All the wooden forms were adjusted

to have round edges for the corners.

Figure ‎3-8: Top view-original shape

Figure ‎3-9: Top view-rounded shape

Each wrapped column had three configuration of wrapping:

The top had only one layer 150mm wrapping width.

The bottom had only one layer 150mm wrapping width.

The middle had either one or two 500mm wrapping width.

40

The reason behind wrapping the top and bottom layer was to remove the stress

concentration and allow for overlapping.

Figure ‎3-10: Column Configuration

3.5 Strain Gauge Fixing

Two strain gauges were used to study the behavior of the column. One was

placed in the longitudinal direction to measure the strain in the vertical direction. And

the other was placed in the transverse direction to measure the strain in the horizontal

direction. Before fixing the strain gauges, the side surface for the entire column was

grinded smoothly to ensure a straight surface when applying the sheet. Both strain

gauges were fixed at the lateral side surface of the concrete column.

Figure ‎3-11: Strain Gauge Fixing

150 mm

500 mm

150 mm

750 mm

41

3.6 CFRP and GFRP Properties

Two types of polymers were used; CFRP and GFRP. The material testing

properties were provided by the manufacturer which can be viewed in the following

table:

Table ‎3-3: Testing Results of FRP materials provided by the manufacturer

Property CFRP GFRP

Thickness (mm) 0.11 0.219

Tensile Strength (MPa) 4800 3400

Tensile Modulus (GPa) 236 35

Ultimate Elongation (%) 2.5 4

3.7 Epoxy Preparation

Epoxy is a thermosetting polymer that cures when mixed with a catalyzing

agent or hardener. It is a versatile polymer with diverse applications such as industry,

painting, coating, adhesives, electronics, and structural and aerospace applications

[37]. In this research, Epoxy is used as a bonding material between the CFRP or

GFRP wraps and the RC short column.

Two types of ingredients were used for the attachment process of the CFRP or

GFRP to the concrete columns; Primer and Saturant. The Primer is a low-viscosity

material used to fill the pores on the concrete specimen surface in order to ensure full

bonding between the FRP composite and the concrete surface. Any source of pores

would decrease the efficiency of the bonding. The primer has two liquid components,

base and hardener. When mixed together it forms a liquid material that is used to fill

the pores in the concrete surface.

Table ‎3-4: Properties of Primer

Property Test Method Value

Component - Base and Hardener

Color - Clear I Pale Yellow

Potlife - 70+/- 10min

Service Temperature - +5 Co to +75 C

o

Surface Drying Time ASTM D2939 6-8 hours

Bond Strength ASTM D4541 Concrete Failure

42

Figure ‎3-12: Primer Part A

Figure ‎3-13: Primer Part B

Figure ‎3-14: Primer Effect [24]

The Saturant is a medium-viscosity material used as a bonding agent between

the concrete surface and FRP material. Upon the application of the primer, the

specimen was left under room temperature for the resin to be dried and cured.

Following this, the saturant was applied on the desired surface with a thickness no

less than 3mm. After the application of saturant, the FRP wraps were applied

immediately before the drying process of the saturant starts. In order to achieve full

bonding between the FRP wraps and the column, it is vital to submerge the FRP in the

epoxy (wet method). Alternatively, an equal amount of success can be found by

attaching the wrap around the column and then applying a second coat of saturant (dry

method).

Table ‎3-5: Properties of Saturant

Property Test Method Value

Component - Base and Hardener

Color - Grey/White/Light Blue

Potlife - 45-60 min

Service Temperature - +5 Co to +75 C

o

Bond Strength ASTM D4541 > 2 N/mm2

Compressive Strength BS 6319-2 70 N/mm2 at 7 days

43

Figure ‎3-15: Saturant part A

Figure ‎3-16: Saturant Part B

3.8 Proposed Matrix

The proposed matrix consists of 30 columns divided into three groups. The

first group has 6 unconfined columns serving as control columns, where 3 were

normal-designed columns and 3 were under-designed columns.

The second group comprises of 12 normal-designed confined columns. 3 of the

12 were normal-design columns wrapped with one layer of CFRP, 3 were normal-

designed columns wrapped with two layers of CFRP, 3 were normal-designed

columns wrapped with one layer of GFRP, and 3 were normal-designed columns

wrapped with two layers of GFRP.

Finally, the last group consists of 12 under-designed confined columns. This

group has 3 under-designed columns wrapped with one layer of CFRP, 3 under-

designed columns wrapped with two layers of CFRP, 3 under-designed columns

wrapped with one layer of GFRP, and 3 under-designed columns wrapped with two

layers of GFRP.

3.9 Columns Designation System

An identification system consists of four characters as “ABCD” was employed

to put a unique name for each confined column. The first character identifies the

design type followed for the column. For example, “N” stands for “normal-design”

which has the minimum number of transverse reinforcements. On the other hand, “U”

stands for “under-design” which has a reduced number of stirrups.

44

The second character identifies the type of material used as a wrapping

material. For example, C stands for “CFRP” and G stands for “GFRP”. Meanwhile

the third character identifies the number of wraps used around the column.

Finally, the last character identifies the serial number for each column starting

from “1” up to “3” since there are three columns of each case. Thus, NG13 will be the

third in the group of normal-designed columns wrapped with one layer of GFRP. It

should be noted that for the unwrapped columns, a two-character designation system

is employed. These two characters identify the design type and the serial number for

each column. For example, “U1” represents the first column in the group of

unconfined under-designed concrete column. The following table summarizes the

entire matrix:

Table ‎3-6: Matrix designation system

N1 N2 N3 U1 U2 U3

NC11 NC12 NC13 UC11 UC12 UC13

NC21 NC22 NC23 UC21 UC22 UC23

NG11 NG12 NG13 UG11 UG12 UG13

NG21 NG22 NG23 UG21 UG22 UG23

3.10 Instrumentation and Testing Procedure

All concrete specimens were subjected to uniaxial compressive load using

Instron 8808 with a capacity of 2400kN. The load was applied at a deflection rate of

0.5mm/min. Prior to testing, all specimens possessed a thick layer of paper attached at

the top and bottom surface of the column in order to ensure that the contact surface

remained parallel and that the applied load remained concentric.

Figure ‎3-17: Testing Equipment

Figure ‎3-18: Specimen under testing

45

CHAPTER 4 DISCUSSION OF EXPERIMENTAL RESULTS

4.1 N Group

4.1.1 Load vs. Deflection

The experimental testing began with the normal-designed columns. The first

group of columns to be tested was N. According to Figure 4-1, the load vs. deflection

curve for N1 starts from zero and travels upwards until it reaches a displacement of

4.430mm. The load continues increasing with a decreased slope value until it reaches

the ultimate point at 5.258mm where the failure load was 938kN.

The load vs. deflection curve For column N2 starts from zero and moves with

an increased slope until it reaches a displacement of 3.391mm. Then, the curve rises

with a decreased slope value till it reaches the ultimate point at 3.975mm. The column

eventually failed at 804kN which makes it closer to the theoretical value.

Columns N3’s load vs. deflection curve rises from zero until it reaches a

displacement of 3.348mm. The curve undergoes a slight decrease in the slope until it

reaches the ultimate point at 4.302mm. The column eventually failed at 843kN.

After looking at all three values of the load, the value of N1 varies greatly

from that of N2 and N3, as is thus considered an anomaly.

46

Figure ‎4-1: N Load vs. Extension Diagram

4.1.2 Stress vs. Strain Diagram

The stress vs. strain curves in for N group is shown in Figure 4-2, the strain

gauge that represents N2 in the axial direction failed to work completely as it

produced no value. As for N1 and N3, the curves produced from the strain gauges

were reasonable and working well. In case of the transverse direction, all columns

produced similar curves thus indicating that the strain gauges have worked properly.

4.1.3 Transverse Strain

The values for each strain were measured based on Figure 4-2. The change in

the transverse strains was -3.47x10-4

for N1, -3.45 x10-4

for N2, and -6.49 x10-4

for

N3. The value for N3 in the transverse direction showed a large difference from other

values and was thus excluded from calculations when trying to ascertain the average.

4.1.4 Ductility Index

All ductility indexes values for the three columns were collected based on the

division of deflection at ultimate load by the deflection at yield load. Consequently, it

was found that N3 gave the highest ductility of 1.285 compared to 1.187 and 1.172,

for N1 and N2 respectively.

47

Figure ‎4-2: N Axial vs. Transverse Strain

4.2 NC1 Group

4.2.1 Mode of Failure

The second group to be tested was “NC1”. What differentiated this group from

N1 was the type of material and number of layers wrapped around the column (i.e.

RC short columns wrapped with one layer of CFRP). As can be seen below, most

columns had a failure at the top of the column due to the delamination of the CFRP

wrap.

Column NC11 had failure due to the delamination of the upper CFRP

confinement. Additionally, an eccentricity occurred in the column due to unbalanced

orientation of the column’s top surface which might not have been completely

horizontal. This buckling caused the middle wrapped layer to split and completely

remove from the column as shown in Figure 4-3 and Figure 4-4.

48

Figure ‎4-3: NC11 Two Sides view

Figure ‎4-4: NC11 One Side view

The failure in NC12 also appeared at the top confinement, most of which was

delaminated leaving only a small portion still attached to the column. However, it

didn’t experience any eccentricity as shown in Figure 4-5 and Figure 4-6.

Figure ‎4-5‎: NC12 corner view

Figure ‎4-6: NC12 One side view

NC13 experienced failure at the top confinement region of the column. The

layer only experienced delamination and the failure extended a few centimeters to

almost the middle of the column according to Figure 4-7 and Figure 4-8.

49

Figure ‎4-7: NC13 Corner View


4.2.2 Load-Extension Diagram

All the graphs for this group were drawn in Figure 4-9. The curve for NC11

starts from zero-load and rises with an increased slope until it reaches a displacement

of 4.765mm. After this point, the load keeps on increasing with a decreased value of

the slope until it reaches the ultimate point at 6.266mm after which the load has

continues to decrease. The column failed at the ultimate load of 917kN.

For NC12, similarly to NC11, its load curve goes from zero and rises until it

reaches a displacement of 4.143mm. Unlike column NC11, the column was unable to

handle more displacement causing it to quickly reach the ultimate point at 4.886mm.

The column failed at ultimate load of 1002kN.

NC13, on the other hand, behaved differently than NC11 and NC12. Here, the

load curve rises from zero-load at an increasing slope until 1.0mm. Then, the curve

decreases in slope making the column handle lesser loads for that amount of

deflection. At a deflection of 2.5mm, the slope increases with a constant rate until it

reaches a displacement of 4.090mm. Beyond this point, the column was capable of

handling more loads until it reached ultimate point at 5.007mm where the column had

a failure load of 910kN.

50

Figure ‎4-9: NC1 Load vs. Extension Diagram

4.2.3 Stress-Strain Diagram

All the strain gauges were operational according to the stress vs. strain curves

in Figure 4-10. The strain gauge for NC13 in the axial direction behaved differently

from the other columns. The transverse direction also had a lower value for the slope

than the other two columns. Looking at NC11 and NC12, the curves share similar

characteristics as they demonstrate identical behavior in axial or transverse direction.


The values for the transverse direction were -2.38 x10-4

for NC12 and -3.10

x10-4

for NC13. The values for NC11 were vastly different from others. Therefore, it

was excluded from the average calculations.

51

Figure ‎4-10: NC1 Axial vs. Transverse Strain


All the ductility indexes values for the yield and ultimate points for the three

columns were collected from the load and deflection curves. NC11 demonstrated the

highest value of 1.135 in comparison to NC12’s 1.179 and NC13’s 1.224.

4.3 NC2 Group


The third group to be tested was NC2. This group differs from NC1 group in

the number of layers wrapped around the column. Here, all columns were wrapped

with two layers of CFRP. As can be observed from the figures below, virtually all

columns shared the same failure type i.e. delamination of the top confinement layer.

NC2 shared certain characteristics with NC1. According to Figure 4-11 and

Figure 4-12, the failure in column NC21 has occurred in the top confinement layer by

the complete removal of the layer and removing a small portion of the middle CFRP

wrap. In addition, the top transverse reinforcement has become visible which

indicates to us that the column has suffered from severe damage due to the applied

load.

52


Figure ‎4-12: NC21 Side View

The failure in NC22 was similar to the failure in NC21 according to

Figure 4-13 and Figure 4-14. The failure occurred in the top part of the column which

caused the top confinement to be delaminated as well. In addition, a small portion was

removed from the middle layer. The column has also suffered a severe damage in the

top part due to the applied load.



The failure in NC23, as shown in Figure 4-13 and Figure 4-14, appeared in the

CFRP confinement as debonding at the top section. What made the failure in this

column different from NC21 and NC22 was the bonded CFRP wrap, which suffered a

severe rupture. However, the confined layer of CFRP remained attached to the

column.

53




All the graphs for the Load vs. Extension diagram in Figure 4-17 share similar

behavior. The curve for NC21 rises steadily from zero and reaches a displacement of

4.212mm. Then, the curve keeps rising with a lower-value slope until hitting ultimate

point at 5.701mm. Beyond failure point, the columns were able to maintain the same

level of load while handling more extension which helps us conclude that the

confinement level was sufficient. The column eventually failed at a load of 1014kN.

The curve for NC22 ascends from zero-load going with a constant slope until

it reaches a displacement of 4.076mm. After that, the slope kept rising until it reaches

the ultimate point at 4.952mm, where the column failed at 1010kN.

The behavior for NC23 was different from the previous two columns. The

curve starts from zero-load moving upwards with a constant slope until it reaches a

displacement of 3.956mm. After that, the column undergoes large displacement for

nearly 4cm while maintaining small changes in the load. This indicates that the

confinement layer was very efficient. After this point, the column reached the ultimate

point at 5.437mm. The column eventually failed at a load of 1019kN.

54

Figure ‎4-17: NC2 Load vs. Extension Diagram


In the case of stress vs. strain curves in Figure 4-18, the strain gauge for the

axial direction for NC21 didn’t work since the strain has not started from zero. For the

transverse direction, it has not worked at all due to damaged wires or improper

attachment to the column. As for the other strain gauges that represent NC22 and

NC23, both demonstrate identical characteristics.


The values for the transverse strain were -5.00 x10-5

for NC22 and -4.00 x10-5

for NC23. The values for NC21 in the transverse direction were omitted due to the

functionality of the strain gauge.

55

Figure ‎4-18: NC2 Axial vs. Transverse Strain


Pertaining to the ductility indexes, the values for the three columns were

calculated based on the division of the displacement at the ultimate load by deflection

at yield load. NC21 and NC23 had the highest values for ductility, 1.353 and 1.374

respectively. NC22, on the flip side, had a comparatively lower value of 1.215.

4.4 NG1 Group

4.4.1 Modes of Failure

The forth group to be tested was NG1. This group represents the columns

designed for the minimum number of transverse reinforcements wrapped i.e. one

layer of GFRP. From the below figures, it can be noted that the mode of failure varies

from one column to another.

The failure in NG11, according to Figure 4-19 and Figure 4-20, occurred in the

form of debonding of the middle confinement layer, not in the top confinement layer

as demonstrated in the previous columns. The failure occurred as a slight rupture in

the middle layer with slight damage in the concrete.

56

Figure ‎4-19: NG11 Corner View

Figure ‎4-20: NG11 Side View

The mode of failure for NG12, as shown in Figure 4-21 and Figure 4-22, has

only appeared as debonding of the top confinement layer of the column. The failure

occurred as a small rupture in the top layer without removing it. In addition, the

column suffered no severe damage. What little damage did occur, took place in the

ruptured side of the concrete.



The mode failure for NG13, according to Figure 4-23 and Figure 4-24, hasn’t

appeared properly. The rupture in the column was not shown neither in the top

confinement layer nor the bottom layer. However, a small crack was indicated in the

middle confinement layer.

57


Figure ‎4-24: NG13 Closer Side View


The curves representing the load vs. extension for each column, according to

Figure 4-25m, are different. For NG11, the curve starts from zero-load and moves

upwards with a constant slope until it reaches a displacement of 4.438mm. The slope

then changes decreases until it reaches ultimate point at 5.639mm where the column

failed at 977kN.

The curve for NG12 moves upwards from zero-load with constant slope until

it reaches a displacement of 5.765mm. Then the curve keeps rising with a lower value

of slope until it reaches ultimate point at 8.147mm. It appears that there is a big

difference between the yield point and ultimate point, indicating that the column was

not only able to handle the increase in extension but also maintain the same level of

load. This was possible due to the strong confinement by the GFRP layer. At the

ultimate load, the column failed at 926kN.

The curve for NG13 rises from zero-load with a constant slope until it reaches

a displacement of 4.786mm. After that, the curve keeps on moving upward with

decreased value for the slope until it hits ultimate point at 6.650mm, where the

column failed at 890kN.

58

Figure ‎4-25: NG1 Load vs. Extension Diagram


All the strain gauges for this group worked properly according to the curves

shown in Figure 4-26. In addition, there was large variation in the axial direction

values. However, as can be seen below, the curves that represent the stress-strain

relationship for the transverse direction share the same characteristics.


The values for each strain gauge were measured based on Figure 4-26. The

change in the transverse strain values were -1.5 x10-3

for NG11, -3.32 x10-4

for NG12,

and -2.50 x10-4

for column NG13. The value for NG11 exhibited a large difference

from the other two values, thus considered the anomaly in these results.

59

Figure ‎4-26: NG1 Axial vs. Transverse Strain


The ductility indexes values for the three columns vary. Column NG12 has

made the highest value for the ductility (1.413) compared to column NG11 (1.271)

and NG13 (1.390).

4.5 NG2 Group


The fifth group to be tested was NG2. This group represented columns that

were wrapped with two layers of GFRP. From the below figures, it is clear that the

mode of failure varies from one column to another.

The mode of failure for column NG21, as shown in Figure 4-27 and

Figure 4-28, was demonstrated as debonding of some part of the top confinement

wrap. The remaining portions the top confinement layer remained attached to the

column.

60


Figure ‎4-28: NG21 Top view

The failure mode in column NG22 was different from column NG21 as shown

in Figure 4-29 and Figure 4-30. The mode of failure was shown as debonding of the

top part of the middle confinement layer.



The failure mode in column NG23, as shown in Figure 4-31 and Figure 4-32,

appeared as a rupture of some part of the top confinement GFRP layer. The column

didn’t exhibit any kind of severe damage on the concrete below the top confinement

layer. This indicated that the column suffered from a debonding failure.

61


Figure ‎4-32: NG23 Different Side View


The load vs. extension curves for the columns in this group, as shown in

Figure 4-33, are almost similar to each other. For column NG21, the curve starts from

zero-load moving upwards with a constant slope until it reaches a displacement of

3.543mm. After that, the curve rises with a lower value for the slope until it arrives at

the ultimate point of 7.178mm. The difference between the yield and ultimate points

is large. This indicates that the column was able to handle large deflection while

maintaining almost the same level of load. The column finally failed at 1044kN.

The curve for NG22 moves upwards from zero with a constant slope, until it

reaches a displacement of 3.139mm. Later on, the curve keeps rising with a lower

value of the slope until it reaches ultimate point at 7.489mm. Column NG22 achieved

the maximum load compared to all other columns in the matrix. This column failed at

1141kN.

The curve for NG23 starts from zero-load and moved upwards with a constant

slope until it reaches a displacement of 4.040mm. After that, the curve goes on rising

before hitting ultimate point at 4.656mm. Here, the difference between the yield and

ultimate point is smaller than column NG23. However, the column maintained an

identical load after passing the ultimate point which can give us an indication of

strong confinement. This column eventually failed at 1035kN.

62

Figure ‎4-33: NG2 Load vs. Extension


The strain gauges curves representing the change in axial and transverse strain,

as shown in Figure 4-34, were working properly. Looking at the curves, the behavior

of column NG22 in the axial and transverse direction varies from the others.


Measurements of the values for each strain were based on Figure 4-34. The

change in the transverse strain was about -5.40x10-5

for column NG21, -6.00 x10-5

for

column NG22, and -4.6 x10-5

for column NG23. The value for column NG23 in the

axial direction was not shown since its strain gauge was out of order.

63

Figure ‎4-34: NG2 Axial vs. Transverse Strain


The ductility index for NG21 and NG22 are close to each other, 2.026 and

2.386 respectively, indicating that these columns were able to withstand that high

level of load due the strong confinement by the GFRP layer. NG23, on the other hand,

had a comparatively lower value of 1.152.

4.6 U Group


The sixth group to be tested was the U group. Basically, U was a set of

columns designed with a number of transverse reinforcements that were lower than

the minimum requirement by the ACI-318 code. The number of transverse

reinforcements was reduced to four stirrups (after it was six stirrups in the normal-

design).

The load vs. extension curve from for column U, according to Figure 4-35,

starts from zero-load moving upwards with a constant slope until it reaches a

displacement of 2.498mm. The curve keeps on going upwards with a lower value of

slope until it hits an ultimate point of 3.084mm, where the column failed at 658kN.

64

The curve for U2 starts from zero-load and moves upwards with a lower slope

value than column U1 until it reaches a displacement of 3.407mm. After that, the

curve keeps on going upwards, eventually reaching ultimate point at 4.329mm. The

column failed at 546kN, considered to be the lowest value in the matrix.

After moving up from zero-load with a constant slope, the curve for U3

reaches a displacement of 4.424mm. After that, the curve rises until it hits ultimate

point at 3.466mm. The curve plummets after this failure point, with the column failing

at 742kN.

Unlike the load results for N group, U group demonstrated lower load results

due to the reduced number of transverse reinforcements.

Figure ‎4-35: U Load vs. Extension Diagram


All the strain gauges, as represented in Figure 4-36, were working properly.

For strain gauges that represented the axial direction, a big variation occurs between

the curves that represent the columns in this group. As for the transverse strain

gauges, all the curves were almost similar in behavior.

65


The values for the transverse strain were -7.50x10-4

for U1, -2.50 x10-4

for U2,

and -3.10 x10-4

for U3. The value for U1 in transverse direction showed a large

difference compared to the other values. For this reason, it was neglected when

calculating the average.

Figure ‎4-36: U Axial vs. Transverse Strain


All ductility indexes values were calculated based on the division of the

deflection at ultimate point by the deflection at yield. Column U1 and column U2

demonstrated closer value, of 1.235 and 1.271 respectively. However, column U3 had

a ductility of 1.012, which was the lowest value for not only this group but the entire

matrix.

4.7 UC1 Group


The sixth group to be tested was UC1. This group comprised of the under-

designed columns strengthened with one layer of CFRP wrap. From the figures

below, the mode of failure is almost the same for all the columns.

66

The mode of failure for UC11, according to Figure 4-37 and Figure 4-38,

appeared in the top part of the column by the partial delamination of the top layer and

some of the upper side of the middle layer. In addition, there was severe damage to

the concrete cover which led to the appearance of the top transverse reinforcement.

Figure ‎4-37: UC11 Corner View

Figure ‎4-38: UC11 Close Corner View

The mode of failure for UC12, as shown in Figure 4-39 and Figure 4-40, took

place at the top part of the column in the form of a full delamination of the top

confinement layer and partial removal in the top part of the middle confinement wrap.

In addition, damage also occurred to the concrete column which led to the occurrence

of transverse reinforcements.



UC13 demonstrated the most severe case when compared to the entire matrix.

According to Figure 4-41 and Figure 4-42, there was full delamination in the top

67

confinement layer and even in portions of the middle layer. In addition, the main

reinforcement bars also buckled slightly.


Figure ‎4-42: UC13 Close Corner View


The curves for the Load vs. Extension diagrams in Figure 4-43 have the same

movement. For column UC11, the curve started from zero-load moving upwards with

a constant slope until it reaches a displacement of 3.702mm. After that, the curve

moves upwards with a lower value for the slope until it reaches an ultimate point of

4.163mm. This shows that the column failed at 871kN.

The load vs. deflection curve for UC12 has a similar trajectory to UC11,

starting from zero-load and moving upwards with a constant slope before arriving at

displacement of 6.745mm. After this point, the curve moves upwards with a lower

value for the slope before reaching an ultimate point of 8.995mm, where the column

failed at 912kN.

On the other hand, UC13 curve starts from zero-load, moving upwards with a

constant slope until it reaches a displacement of 5.342mm. After that, the curve keeps

on moving upwards with a lower value for the slope until it reaches ultimate point at

6.245mm at which point the column failed at 893kN.

68

Figure ‎4-43: UC1 Load vs. Extension


There was only one strain gauge to represent the change in the axial strain as

shown in Figure 4-44. This strain gauge belonged to UC13. Pertaining to transverse

direction, UC12 and UC13 appeared to have an identical demonstration for the

change in the transverse direction.


The values for the transverse strain were -7.40x10-4

for column UC11, -2.33

x10-4

for column UC12, and -1.90 x10-4

for column UC13. Column UC11 showed a

big difference in comparison to other values for the transverse strain gauges and was

thus omitted when calculating the average.

69

Figure ‎4-44: UC1 Axial vs. Transverse Strain


The ductility indexes values for the three columns seem to increase as the load

increase. For UC11 the value for the ductility is 1.125 and the corresponding load is

871kN. For UC12, the value for the ductility is 1.334 and the corresponding load is

912kN. And lastly, for column UC13, the value for the ductility is 1.169 and the

corresponding load is 893kN

4.8 UC2 Group


The eighth group to be tested was UC2. This group differs from UC1 in the

number of layers wrapped around the column, that is, two layers of CFRP. As

indicated by the figures below, almost all the columns shared the same type of failure

which was the delamination of the top confinement layer. UC2 possessed similar

characteristics to those of UC1 except that the failure occurred only in the upper

confinement.

The failure in UC21 has happened in the top confinement layer by the

delamination of some portions of the top layer while remaining portions were attached

70

to the column. In addition, part of the transverse reinforcement also appeared due to

the partial delamination. The change in the column condition can be seen in

Figure 4-45 and Figure 4-46

Figure ‎4-45: UC21 Side View

Figure ‎4-46: UC21 Close Side View

Pertaining to Figure 4-47 and Figure 4-48, the failure in UC22 is similar to the

failure in UC21. It appeared in the top section of the column which caused part of the

top confinement to be delaminated while the remaining part remained attached. Also,

part of the transverse reinforcement also appeared due to the partial delamination.


Figure ‎4-48: UC22 Top View

The failure in column UC23 also appeared in the CFRP confinement at the top

section of the column. The failure manifested as delamination of the top confinement.

Unlike the previous two columns, this column suffered delamination without the

appearance of the main or transverse reinforcement. However, it is clear in the figures

71

that the concrete cover has also suffered from minor damage. All these effects can be

seen in Figure 4-49 and Figure 4-50.


The Load vs. Extension curve in Figure 4-51 shows that the curves

representing load vs. extension for each column in this group, differ from each other.

The curve for UC21 starts from zero-load, moves up with a constant slope until it

reaches a displacement of 6.706mm. Later on, the slope changes direction with a

decrease in its value. This continues until it reaches ultimate point at 7.655mm.

Beyond this point, the curve goes down indicating that the column has failed at

1035kN.

The curve for UC22 behaves differently. From zero-load, it increases with a

constant slope until it reaches a displacement of 5.131mm. Then, the curve keeps on

rising with a lower value of slope, finally reaching ultimate point at 6.629mm at

which point the column failed at 1053kN.

The curve for UC23 goes upwards from zero-load with a constant slope until it

reaches a displacement of 5.362mm. After that, the curve keeps on moving upward

with decreased value for the slope until reaching the ultimate point at 6.437mm. The

column has failed at 1035kN. Thus, the load values for each column are somehow

related to each other.


Figure ‎4-50: UC23 Side View

72

Figure ‎4-51: UC2 Load vs. Extension


In the case of the stress vs. strain curves in Figure 4-52, the strain gauge that

represents that strain change in the axial direction for column UC21 didn’t function

properly due to irregular strain behavior. UC22 and UC23 are almost similar in the

behavior. For the transverse direction, all the strain gauges operated very efficiently.


The values for each strain were measured based on The change in the axial

strain was roughly 1.021 x10-2

for UC21, 8.838 x10-3

for UC22 and 8.583 x10-3

for

UC23. In regards to transverse strain, the values were about -4.19x10-4

for UC21, -

3.68 x10-4

for UC22 and -1.00 x10-4

for UC23.

73

Figure ‎4-52: UC2 Axial vs. Transverse Strain


In terms of ductility indexes, the values for the three columns increased as the

load increased. UC21 demonstrated that the value for the ductility of 1.142 and the

corresponding load as 1035kN. UC22 depicted ductility equal to 1.292 and a

corresponding load of 1053kN. Column UC13, similarly, possessed ductility value of

1.200 and a corresponding load of 1035kN

4.9 UG1 Group


The ninth group to be tested was UG1. This group represented the under-

designed columns strengthened with only one layer of GFRP wrap. As depicted by the

figures below, the mode of failure varied from one column to another.

The failure in UG11, according to Figure 4-53 and Figure 4-54, took place as

debonding of the top layer and portions of the middle confinement layer. However,

the concrete cover suffered comparatively superficial damage.

74

Figure ‎4-53: UG11 Corner View

Figure ‎4-54: UG11 Different Corner View

The mode of failure for column UG12 occurred in the form of debonding of

some middle portions of the confinement layer. The concrete cover, however, suffered

from minor damage. The changes can be seen in Figure 4-55 and Figure 4-56


Figure ‎4-56: UG12 Side View

The mode of failure for UG13, as shown in Figure 4-57 and Figure 4-58,

occurred in the form of delamination on some of the top confinement layer. The other

parts, on the other hand, remained attached to the column. In addition, the concrete

cover suffered only minor damage.

75




The curves for each column in the load vs. extension curve in Figure 4-59 for

this group relate to each other. Starting from the first column UG11 in this group, the

curve starts from zero-load moving upwards with a constant slope until it reaches a

displacement of 4.890mm. The slope then drops, decreasing in its value until hitting

an ultimate point at 5.529mm. After that point, the curve keeps on decreasing. The

column failure occurred at 845kN.

In the case of UG12, the behavior of the column is similar to UG11. It strarts

from zero-load and moves upwards with a constant slope until it reaches a

displacement of 5.023mm. Then, the curve rises with a lower value of slope before

reaching ultimate point at 6.523mm. This demonstrates that the column was able to

handle the increased deflection by maintaining minor change in the load due to the

strong confinement by the GFRP layer. At the ultimate point, the column failed at

892kN.

The UG13 curve goes from zero-load and continues moving upwards with a

constant slope before reaching a displacement of 5.826mm. After that, the curve

maintains an upward movement with a decreased value for the slope. This continues

until it reaches ultimate point at 7.612mm. The column eventually failed at 831kN.

76

Figure ‎4-59: UG1 Load vs. Extension


The strain gauge representing strain change in the axial direction for all the

columns failed to work at all as shown in Figure ‎4-60. For transverse direction,

UG11 strain gauge failed to work as well. The strain gauge for the remaining columns

appears to be working and behaves in identical fashion.


The values for the transverse strain gauges were -5.00 x10-4

for UG12 and -

3.00 x10-4

for UG13. For UG11 transverse strain gauge, the value is unavailable due

to the breakdown of the strain gauge.

77

Figure ‎4-60: UG1 Axial vs. Transverse Strain


The ductility indexes values were calculated based on the division of

deflection at ultimate point by the deflection at yield. Almost all columns had relevant

ductility indexes. The difference in ductility index between UG12 (1.299) and UG13

(1.307) was minimal. UG13 achieved a ductility index of 1.131 making it greater than

the previous two ductility indexes.

4.10 UG2 Group


The tenth and last group to be tested was UG2. This group represented the

under-designed columns wrapped with two layers of GFRP. From the figures below,

the mode of failure varied from one column to another. From Figure 4-61 and

Figure 4-62, the mode of failure for UG21 was demonstrated as delamination of the

top confinement layer. Also, the concrete cover suffered minor damage when the

GFRP wrap was delaminated.

78



The failure mode in UG22, as shown in Figure 4-63 and Figure 4-64, occurred

during debonding of the top and some of the middle confinement layer, while the

other part remained attached to the column. In addition, the concrete cover only

suffered from minor damage.



The failure mode in UG23, according to Figure 4-65 and Figure 4-66, occurred

during debonding of the top confinement layer. In addition, the concrete cover only

suffered minor damage.

79




The curves for each column in the load vs. extension in Figure 4-67 for this

group were different in behavior. Starting from UG21, the curve starts from zero-load

moving upwards with a constant slope until it reaches a displacement of at 4.446mm.

The slope then drops in value before reaching ultimate point at 5.601mm. After that

point, the curve keeps decreasing. Column failure occurred at 938kN.

The behavior of UG22 curve is identical to that of UG21, starting from zero-

load, moving upwards with a constant slope until it reaches a displacement of

7.598mm. Then, the curve rises with a lower value of slope until hitting ultimate point

at 9.401mm. This indicates that the column was able to handle the increased

deflection by maintaining a minor change in the load due to the strong confinement by

the GFRP layer. The column failed at 940kN.

Finally, for UG23 the curve starts from zero-load and moves upwards with a

constant slope until it reaches a displacement of 4.478mm. After that, the curve

continues ascending with decreased value for the slope until it reaches ultimate point

at 6.014mm. Column failure takes place at 997kN.

80

Figure ‎4-67: UG2 Load vs. Extension


In terms of the stress vs. strain curves in Figure 4-68, the strain gauge

representing UG23 in the axial and transverse direction failed to work at all. As for

the other columns, the behavior of the strain in the axial direction for both columns

differed from each other. For the transverse direction, they appeared virtually

identical.


The values for each strain was measured based on Figure 4-68. The change in

transverse strain values were -3.19 x10-4

for UG21 and -1.66 x10-4

for UG22. For

UG23 transverse strain gauge, the value is unavailable due to the breakdown of the

strain gauge.

81

Figure ‎4-68: UG2 Axial vs. Transverse StrainDuctility Index

Pertaining to ductility indexes, all values were calculated based on the division

of the deflection at ultimate point by the deflection at yield. All columns had relevant

ductility indexes. The difference in ductility index between UG22 (1.237) and UG23

(1.260) was minimal. As for UG21, it has achieved a ductility index of (1.343), which

was higher than the previous two ductility indexes.

82

CHAPTER 5 THEORETICAL RESULTS

5.1 Introduction to ACI 440.2R

In the past few decades, several researchers have conducted experiments in

order to investigate the behavior of the FRP materials in strengthening structural

members. Such studies were necessary in understanding the behavior of these

materials so that equations (and parameters) could be designed that accurately

predicted desired characteristics. American Concrete Institute (ACI) committee

devised a design code manual containing equations that predict the load for beams

and columns externally strengthened with FRP materials. This code was referred to as

ACI 440.2R.

5.2 Design Equations

One of the cases mentioned in this code, talks about how to predict the

confined compressive strength of non-circular concrete columns when subjected to

axial load without bending moment. The theoretical compressive strength value of the

non-confined concrete column was calculated based on Equation (1). However, when

confinement is taken into consideration, Equation (1) had to be modified to include

the compressive strength of confined concrete. That is, the concrete compressive

strength ( ) was replaced with (

) to form Equation (2).

[ ( ) ]-----------------------------------------------Equation 2

In order to predict the theoretical load for confined concrete column, ACI 440

suggests the following five steps to be followed:

Step 1: Compute the design FRP material properties:

Some of the properties provided by the manufacturer, such as the ultimate

rupture strain ( ), do not represent the material subjected to environmental

condition. Therefore, an environmental reduction factor ( ) should be applied to

account for this matter. This factor has different value from one material to another.

For example, Carbon has a value of 0.95 while Glass possesses a value of 0.75. To

calculate the effective strain level ( ) and design rupture strain ( ), Equation 3 and

(4) are to be followed:

83

----------------------------------------------------------------------------Equation 3

---------------------------------------------------------------------------Equation 4

In Equation 3, ( ) refers to efficiency factor equal to 0.55 for FRP strain to

account for the difference between observed rupture strain in confinement and rupture

strain determined from tensile tests

Step 2: Compute the unconfined column properties:

Such properties include calculating the gross area of the column , the area of

the steel reinforcement ( ), and the ratio of the steel area to the gross area .

Step 3: Compute the confined column properties:

Testing showed that confining square and rectangular members with FRP

jackets can provide marginal increases in the maximum axial compressive strength

( ) of the concrete member [38-42] . In addition, ACI 440 code provisions are not

recommended for aspect ratio (h/b) to exceed 2.0 or dimensions that are larger than

900mm. This was applied when calculating the ratio of the affective area ( ) to

concrete area ( )

⌊( )( )

( )( )

⌋

----------------------------------------------------Equation 5

The previous parameter was used in calculating the shape factor ( ).

Basically, the shape factor depends on two parameters: the cross-sectional area of

effectively confined concrete ( ), and the dimension aspect ratio ( ⁄ ) of the

member

(

)

---------------------------------------------------------------------------Equation 6

Step 4: Compute the maximum confining pressure due to the FRP jacket:

The diagonal for non-circular cross-sectional members was set to be

equivalent to the diameter of circular cross-sectional member as shown in Figure 5-1.

The diameter was calculated according to the following equation:

84

√ -------------------------------------------------------------------------Equation 7

Figure ‎5-1: Equivalent circular section

The maximum confinement pressure ( ) and the maximum confined concrete

compressive strength ( ) were calculated using Equation 8 and 9 [43] [44], with the

inclusion of an additional reduction factor ψf = 0.95 and where (n) represents the

number of confinement layers. The value of this reduction factor was based on the

ACI 440 committee’s judgment.

--------------------------------------------------------------------------Equation 8

----------------------------------------------------------------Equation 9

Step 5: Compute the theoretical compressive load for the non-confined and confined

concrete column:

To do this, Equation 1 and 2 were utilized to calculate the theoretical load.

Equation 1 calculated the non-confined compressive load of the concrete column

while Equation 2 calculated the confined compressive load of the concrete column.

Lastly, a percentage increase was derived from the findings of the two loads to

calculate how much the load increased. The following table illustrates how the

calculations were handled in the excel sheet:

b

85

Table ‎5-1: Excel Sheet for the calculations of the confined load

5.3 ACI Prediction

5.3.1 CFRP Confinement

With the application of one-layer of CFRP, the confined concrete compressive

load increased to 935kN, a percentage difference of 12.75% from unconfined column.

For the application of two-layers of CFRP, the confined concrete compressive load

increased to 1041kN, a percentage difference of 25.5% from unconfined column.

# 1 - εfu 0.03 mm/mm

tf 0.352 mm εfe 0.017 -

ffu* 3400 N/mm2

Ef 55 KN/mm2

εfu* 4.00% - Ag 22500.00 mm2

As 314.16 mm2

ρs 1.40% -

b 150 mm

d 150 mm

rc 25 mm Ae/Ac 0.699508 -

Φ 1 - Ka 0.699508 -

f'c 45 MPa fl 2.3 Mpa

fl/fc 0.1 >0.08

fcc' 50.2 Mpa

# 4 -

size 10 mm

Es 200 KN/mm2 Pn(old) 830 KN

fy 600 MPa Pn(new) 908 KN

% Incr. 9.48% -

CE 0.75 -

FRP Properies

Concrete Properties

Env. Reduction Factor

STEP 1

STEP 4

STEP 5

Steel Properties

STEP 2

STEP 3

Beam Specification

86

Table ‎5-2: Theoretical Calculations-CFRP confined column

Property CFRP One layer CFRP Two layers

Design Rupture Strain (εfu) 0.02375 mm/mm

Effective Strain Level (εfe) 0.013 mm/mm

Concrete Gross Area (Ag) 22500.00 mm2

Steel Reinforcement Area (As) 314.16 mm2

Steel Reinforcement Ration (ρs) 1.40%

Effective Confined Area to Concrete Area (Ae/Ac) 0.699508

FRP Reinforcement Efficiency Factor (Ka) 0.699508

FRP Confinement Pressure (fl) 3.037 MPa 6.074 MPa

FRP Confinement Pressure to Concrete

Compressive Strength (fl/fc) 0.067 0.135

Confined Concrete Compressive Strength (f’cc) 52.011 MPa 59.022 MPa

Nominal Load Before Confinement Pn (old) 830 kN

Nominal Load After Confinement Pn (new) 935 kN 1041

% Increase 12.75% 25.50%

5.3.2 GFRP Confinement

With the application of one-layer of GFRP, the confined concrete compressive

load increased to 908kN, making a 9.48% percentage difference from unconfined

column. This increase is considered lower than what CFRP can achieve due to the

properties of the materials. For the application of two-layers of GFRP, the confined

concrete compressive load increased to 987kN, a 18.96% percentage difference from

unconfined column. This increase is considered lower than what CFRP can achieve

due to the properties of the materials.

87

Table ‎5-3: Theoretical Calculations-GFRP confined column

Property GFRP One layer GFRP Two layers

Design Rupture Strain (εfu) 0.030 mm/mm

Effective Strain Level (εfe) 0.017 mm/mm

Concrete Gross Area (Ag) 22500.00 mm2

Steel Reinforcement Area (As) 314.16 mm2

Steel Reinforcement Ration (ρs) 1.40%

Effective Confined Area to Concrete Area (Ae/Ac) 0.699508

FRP Reinforcement Efficiency Factor (Ka) 0.699508

FRP Confinement Pressure (fl) 2.3 MPa 4.5 MPa

FRP Confinement Pressure to Concrete

Compressive Strength (fl/fc) 0.05 0.1

Confined Concrete Compressive Strength (f’cc) 50.2 MPa 55.4 MPa

Nominal Load Before Confinement Pn (old) 830 kN

Nominal Load After Confinement Pn (new) 908 kN 987

% Increase 9.48% 18.96%

5.4 Experimental vs. Theoretical Results

After predicting the confined compressive load for the normal-design

columns, a comparison table was created to measure the difference against the

experimental results. The following equation was employed in calculating the

percentage difference:

88

Table ‎5-4: Theoretical vs. Experimental Results (Percentage Difference For N-Group)

Type N NC1 NC2 NG1 NG2

Theo. Load (KN) 830 935 1041 908 987

Exp. Load (KN) 862 943 1073 931 1014

% Difference 3.81% 0.85% 3.10% 2.54% 2.73%

The percentage difference for the theoretical vs. experimental values for all the

cases was less than 5% which is considered acceptable. Along with the percentage

difference, the percentage increase was also calculated to measure the level of

enhancement contributed by the FRP wraps. The following equation was used to carry

out the calculation process, summarized in Table 5-5:

Table ‎5-5: Theoretical vs. Experimental Results (Percentage Increase for N-Group)


Theoretical Load (KN) 830 935 1041 908 987

% Increase (Theoretical) - 12.75% 25.50% 9.48% 18.96%

Experimental Load (KN) 862 943 1073 931 1014

% Increase (Experimental) - 9.44% 24.56% 8.06% 17.68%

In the previous table, the experimental values are larger than the theoretical

value due to the duration of the curing. All the columns were exposed to a longer

period of curing (longer than 28 days) due to the repairing process of the testing

machine which took some time to be fixed.

For U-Group, the percentage difference calculation process was not applicable

due to the lack of supporting codes for finding the theoretical value.

89

CHAPTER 6 ANALYTICAL MODEL

6.1 Existing Models of FRP-Confined Concrete

In the last decade, various researchers have investigated the behavior of FRP

systems, leading to the development of numerical models that can describe this

behavior. The majority of the confined models have adopted the concept of Richart et.

al [45]:

Based on their experimental results on circular, square, and rectangular CFRP-

confined concrete specimens, Shehata et al. [46] proposed the following equation for

square columns:

( )

Kumutha et al. [30] suggested a similar model, employing a confinement

coefficient of 0.93. The value of k1 is determined depending on a three aspect ratio

(a/b=1.0, 1.25 and 1.66) of reinforced concrete square and rectangular columns

strengthened by GFRP composite. This model is given by:

( )

6.2 Prediction of a New Model

In order to predict a new model, four cases should be identified to determine

the equations representing each of the four cases. These cases are:

Normal-design with CFRP (NC)

Normal-design with GFRP (NG)

Under design with CFRP (UC)

Under design with GFRP (UG).

In predicting an equation for a specific model, three values have to be

considered: unconfined concrete compressive strength ( ), confined concrete

compressive strength ( ), and the compressive strength contributed by the FRP wrap

( ). Two ratios also have to be calculated, which are strengthening ratio and

90

confinement ratio. These data allow us to mark points on a graph from which an

equation can be linearized to represent this case. The following table represents the

confinement and strengthening ratios for all the cases:

Table ‎6-1: Analysis of the experimental results

Normal Column fl/fco fcc/fco Under Column fl/fco fcc/fco

NC11 0.077 1.065 UC11 0.110 1.344

NC12 0.090 1.162 UC12 0.133 1.406

NC13 0.086 1.056 UC13 0.098 1.376

NC21 0.155 1.176 UC21 0.221 1.596

NC22 0.181 1.172 UC22 0.266 1.623

NC23 0.172 1.182 UC23 0.196 1.595

NG11 0.092 1.134 UG11 0.132 1.304

NG12 0.108 1.075 UG12 0.159 1.375

NG13 0.103 1.033 UG13 0.117 1.282

NG21 0.185 1.212 UG21 0.263 1.446

NG22 0.215 1.325 UG22 0.317 1.450

NG23 0.205 1.201 UG23 0.233 1.538

91

6.2.1 Normal Design with CFRP

Figure ‎6-1: NC Analytical Model

[ ( )]

6.2.2 Normal Design with GFRP

Figure ‎6-2: NG Analytical Model

[ ( )]

92

6.2.3 Under Design with CFRP

Figure ‎6-3: UC Analytical Model

[ ( )]

6.2.4 Under Design with GFRP

Figure ‎6-4: UG Analytical Model

[ ( )]

93

6.3 Verification of the New Models

From the work represented by Kumutha el. al. and Shehata et. al., a

verification process was developed to check how close the experimental values are to

values represented by previous researches. In Table 6-2, the model representing

normal-designed columns wrapped with CFRP has minor difference compared to the

values in the work done by Shehata et. al (0.984 for one layer of CFRP and 0.970 for

two layers of CFRP). However, for normal-designed columns wrapped with GFRP,

the difference in calculations is higher than the CFRP wrapped columns (1.062 for

one layer GFRP and 1.121 for two layers GFRP).

Table ‎6-2: Predicted values comparisosns

Theoretical Models fco t fl Group f’cc f’cc/fco Old/New Model

Shehata et. al.

38.30

0.11 3.20 NCI 41.015 1.071 0.984

0.22 6.40 NC2 43.735 1.142 0.970

Karaghool 0.11 3.20 NC1 41.687 1.089

0.22 6.40 NC2 45.079 1.177

Theoretical Models fco t fl Group fcc fcc/fco Old/New Model

Kumutha el. al.

38.30

0.352 3.82 NG1 41.848 1.093 1.062

0.704 7.64 NG2 45.400 1.186 1.121

H. Karaghool 0.352 3.82 NG1 39.403 1.029

0.704 7.64 NG2 40.511 1.058

94

CHAPTER 7 SUMMARY OF RESULTS

7.1 Summary of the Lab Work

A complete matrix of 30 reinforced concrete columns was prepared in the lab.

These columns were divided into two groups. The first group was designed to have

the minimum number of transverse reinforcement while the other group was designed

to have 50% less transverse reinforcement (less than the minimum requirement by the

ACI-318 code).

Each group was further divided into five categories: three columns were

treated as control columns, three columns were wrapped with one layer of CFRP,

three columns were wrapped with two layers of CFRP, three columns were wrapped

with one layer of GFRP, and the remaining three columns were wrapped with two

layers of GFRP.

Upon casting, all edges were rounded in order to remove the stress

concentration and make the confinement as strong as possible. Following the edge

rounding process, epoxy resins were made and, lastly, all wraps were attached to the

column to be prepared for testing.

7.2 Summary of the Mode of Failure

The previous discussions produced a summary of the mode of failure. Some

groups have one mode of failure while others have more. It was also noted that

debonding failure occurred in NG1 and NG2, while delamination failure occurred in

NC1, UC1 and UC2. The remaining groups suffered from both types of failure. The

site of failures was top of the column, middle of the column, or both. Nevertheless,

the most frequent failure occurred at the top of the column.

95

Table ‎7-1: Summary of Mode of Failure

Column Type of Failure Position Column Type of Failure Position

NC11 Delamination Top UC11 Delamination Top + Middle



NC21 Delamination Top UC21 Delamination Top

NC22 Delamination Top UC22 Delamination Top

NC23 Debonding Top UC23 Delamination Top

NG11 Debonding Middle UG11 Debonding Top + Middle

NG12 Debonding Top UG12 Debonding Middle

NG13 Debonding Middle UG13 Delamination Top

NG21 Debonding Top UG21 Delamination Top

NG22 Debonding Middle UG22 Debonding Top + Middle

NG23 Debonding Top UG23 Debonding Top

7.3 Summary of Load and Ductility Results

7.3.1 N Group

For N, wrapping with one or two layers of CFRP saw an increase in the load

by 9.44% and 24.56%. In comparison, one or two layers of GFRP saw lesser increases

of 8.06% and 17.68%, showing that CFRP produce better results in increasing the

load-capacity.

Ductility increased when wrapped with one or two layers CFRP, by 2.06% and

8.15% respectively. However, if wrapped with the same number of GFRP layers, the

increase was significantly higher: 11.77% and 52.67%. This proves that the GFRP

wraps produce higher levels of deflection capacity than CFRP wraps.

96

Table ‎7-2: N Group Load and Ductility Results


Testing Load (KN) 862 943 1073 931 1014

Strength Increase - 9.44% 24.56% 8.06% 17.68%

∆u (mm) 4.512 5.386 5.363 6.812 6.441

∆y (mm) 3.723 4.332 4.081 4.438 3.574

μ∆ 1.215 1.240 1.314 1.358 1.855

Ductility Increase - 2.06% 8.15% 11.77% 52.67%

7.3.2 U Group

The load increased for U when wrapped with one or two layers of CFRP, by

37.54% and 60.49% respectively. When wrapped with one or two layers of GFRP, the

load increased by 32.02% and 47.80%. From the load results, it is clear that CFRP

wraps are more beneficial in increasing the load capacity. However, the enhancement

level in U was lower than N, possibly due to the reduced number of transverse

reinforcement.

Ductility, too, increased when wrapped with one or two layers CFRP, by

3.07% and 3.24% respectively. However, when wrapped with the same number of

GFRP layers, the increase was 6.14% and 9.12%. This is a clear indication that when

it comes to increasing deflection capacity, GFRP wraps are superior to CFRP wraps.

It should also be noted that the level of enhancement achieved was less than that of N.

Table ‎7-3: U Group Load and Ductility Results

Type U UC1 UC2 UG1 UG2

Testing Load (KN) 649 892 1041 856 959

Strength Increase - 37.54% 60.49% 32.02% 47.80%

∆u (mm) 3.626 6.468 6.907 6.555 7.005

∆y (mm) 3.109 5.263 5.733 5.246 5.507

μ∆ 1.173 1.209 1.211 1.245 1.280

Ductility Increase - 3.07% 3.24% 6.14% 9.12%

97

7.3.3 N vs. U Group

A comparison between the results of un-strengthened designed columns and

those strengthened under-designed columns showed that load capacity increased when

wrapped with one or two layers of CFRP. The increase in load capacity was 3.52%

and 20.80% respectively. Furthermore, wrapping with one or two layers of GFRP

decreased load capacity of N by 0.63% and increased that of U by 11.25%

respectively. We can thus conclude that all types of enhancements made up the loss of

load capacity due to the reduction of the transverse reinforcements.

Ductility, similarly, increased when wrapped with one or two layers CFRP,

which was similar to that of the normal control column. It only decreased by 0.49%

and 0.33% for UC1 and UC2 respectively. However, in the case of the UG1 and UG2

group, it increased by 2.47% and 5.35%.

Table ‎7-4: N vs. U Group Load and Ductility Comparisons

Type N UC1 UC2 UG1 UG2

Testing Load (KN) 862 892 1041 856 959

Strength Increase - 3.52% 20.80% -0.63% 11.25%

∆u (mm) 4.512 6.468 6.907 6.555 7.005

∆y (mm) 3.723 5.263 5.733 5.246 5.507

μ∆ 1.215 1.209 1.211 1.245 1.280

Ductility Increase - -0.49% -0.33% 2.47% 5.35%

7.4 Summary of Transverse Strain results

7.4.1 N Group

The transverse strain gauges decreased when wrapped with one or two CFRP

layers. It decreased by -38.70% and -89.93% for NC1 and NC2. As for GFRP wraps,

the strain decreased by -34.89% and -88.07% for NG1 and NG2, respectively.

Table ‎7-5: N group Transverse strain comparisons


Transverse -3.46E-04 -2.74E-04 -4.50E-05 -2.91E-04 -5.33E-05

- -20.809% -86.994% -15.896% -84.586%

98

7.4.2 U Group

The transverse strain decreased when wrapped with one or two layers of

CFRP, by 38.45% for UC1 and 5.60% for UC2, respectively. As for GFRP wraps, the

strain decreased by 42.86% and -13.39% for UG1 and UG2.

Table ‎7-6: U group Transverse strain comparisons

Type U UC1 UC2 UG1 UG2


- 38.45% 5.59% 42.85% -13.39%

7.4.3 N vs. U Group

The transverse strain demonstrated high results compared to normal non-

strengthened columns. When wrapped with one or two CFRP layers, the change in

strain values was 12.04% for UC1 and -14.55% for UC2, respectively. As for GFRP

wraps, the change in strain was 15.61% for UG1 and -29.91% for UG2.

Table ‎7-7: N vs. U group Transverse strain comparisons

Type N UC1 UC2 UG1 UG2


- 12.04% -14.55% 15.61% -29.91%

99

Conclusion

Experimental results for short reinforced concrete column wrapped with CFRP

or GFRP materials were presented in this research where square concrete specimens

were also confined with either carbon or glass fiber wraps. This was done in order to

examine the effects of confinement on load and ductility. Upon analysis and

monitoring the results, the following conclusions were drawn:

External confinement with CFRP or GFRP materials significantly increases

the load and ductility of the normal-design specimen under axial loading.

The results of materials tested demonstrated that CFRP materials produce the

largest lateral confinement pressure to column specimens. However, GFRP

materials produce a higher enhancement in ductility.

Excessive confinement leads to sudden and destructive compressive failures,

which are to be avoided at all costs.

Externally confined concrete column can potentially undergo large

deformations without complete failure.

And finally, external confinement enhances the loss in the transverse strain.

Proposed Future Work

With the significant outcomes derived from this research, there is potential for

an even more beneficial implementation of these outcomes through an expansion of

the research area. To do so, the following suggestions ought to be taken into

consideration:

Investigating the contribution of AFRP, steel and aluminum when defined as a

wrapping material.

Replacing the transverse reinforcement with aluminum reinforcement.

Increasing the radius of the edge-curved columns in order to relate it with the

compressive strength.

100

References

[1] GangaRao, H. V. S., Taly, N., and Vijay, P. V., Reinforced Concrete Design with FRP

Composites, New York: CRC Press, 2007, pp. 85-91.

[2] [Online]. Available: http://www.fibermaxcomposites.com/shop/images/K300T2.jpg.

[3] [Online]. Available: https://carbonsales.com/images/C/%2361052-50.jpg.

[4] [Online]. Available:

http://www.rubbersealing.com/TCI/images/upload/Image/Texturized%20Glass%20Fiber

%20Cloth.jpg.

[5] American Concrete Institute, ACI 440.2R-08: Guide for the Design and Construction of

Externally Bonded FRP Systems for Strengthening Concrete Structures, Farmington Hill,

MI: ACI Committee 440, 2008.

[6] Adams, P., "Glass corrosion A record of the past? A predictor of the future?," Journal of

Non-Crystalline Solids, vol. 67, no. 1-3, pp. 193-205, September 1984.

[7] Springer, G.S., Environmental effects on composite materials, 3rd ed., Lancaster, PA:

Technomic Publishing Company, 1988.

[8] Parkyn, B., Encyclopedia of Polymer Science and Engineering, 2nd ed., London: John

Wiley & Sons, 1985.

[9] Rao, R.M.G.K.; Balasubramanian, N.; Chanda, M., "Factors Effecting Moisture Absorption

in polymer composties. Part II: Inluence of External Factors," Journal of Reinforced

Plastics and Composites, vol. 3, no. 3, pp. 246-253, July 1984.

[10

]

Allred, R.E., "The Effect of Temperature and Moisture Content on the Flexural Response

of Kevlar/Epoxy Laminates: Part II. [45/0/90] Filament Orientation," Environemental

Effects on Composite Materials, pp. 100-116, March 1984.

[11

]

Devalapura, R.K.; Greenwood, M.E.; Gauchel,, J.V.; Humphrey, T.J., "Evaluation of GFRP

performance using accelerated test methods," in Proc. CDCC 98, Québec, Canada, 1998.

[12

]

Katz, A.; Berman, N.; Bank, L.C., "Effect of cyclic loading and elevated temperature on

the bond properties of FRP rebars," in Proc. CDCC 98, Québec, Canada, 1998.

[13

]

Prichard, G.; Speake, S.D., "Effects of temperature on stress-rapture times in

glass/polyester laminates," Composites, vol. 19, no. 1, pp. 29-35, 1988.

[14 GangaRao, H.V.S.; Vijay, P.V.; Gupta, R.K.; and Barbero, E, "Mechanical-hygrotheramal

101

] responses and predictive models for FRP composites," 2001.

[15

]

Duta, P.K.; Hui, D.; Prasad, Y., "Influence of subfreezing temperatures on the flexural

behavior of thick composites," in 2nd Biennial European Joint Conf. Eng. Sys. Des. Analy.

(ESDA), London, England, 1994.

[16

]

Ballinger, C.; Maeda, T.; Hoshijima, T., "Strengthening of Reinforced Concrete chimneys,

columns and beams with carbon fiber reinforced plastics," in International Symposium

on Fiber-Reinforced-Plastic Reinforcements for Concrete Structures, 1993.

[17

]

Demers, M.; Neale, K. W., "Confinement of Reinforced concrete columns with fiber-

reinforced composites sheets-an experimental study," Canadian Journal of Civil

Engineering, vol. 26, pp. 226-241, 1999.

[18

]

Fardis, M.N.; Khalili, H., "Concrete encased in fiberglass-reinforced plastic," American

Concrete Institute (ACI), vol. 78, no. 6, pp. 440-446, 1981.

[19

]

Fardis, M.N.; Khalili, H., "FRP-encased concrete as a structural material," Magazine of

Concrete Research, vol. 34, no. 121, pp. 191-202, 1982.

[20

]

American Concrete Institute, ACI 440R-96: State-of-the-art report on fiber reinforced

plastics (FRP) reinforcement for concrete structures, Michigan, USA: Farmington Hills,

1996.

[21

]

Nanni, A; Norris, M. S, "FRP jacketed concrete under flexure and combined flexure-

compression," Construction and Building Materials, vol. 9, no. 5, pp. 273-281, 1995.

[22

]

Ohno, S.; Miyauchi, Y.; Kei, T.; Higashibata, Y., "Bond properties of CFRP plate joint," in

Non-Metallic (FRP) Reinforcement for Concrete Structures, Proceeding of the Third

International Symposium, Sapporo, Japan, 1997.

[23

]

Xiao, Y.; Ma, R., "Seismic retrofit of RC circular columns using prefabricated composite

jacketing," Journal of Structural Engineering, ASCE, vol. 123, no. 10, pp. 1357-1364,

1997.

[24

]

Teng, J. G.; Chen, J. F.; Smith, S. T.; Lam, L., FRP-Strengthened RC Structures, West

Sussex, England: Wiley, 2002.

[25

]

Mirmiran, A.; Shahawy, M.; Samaan, M.; El Echary, H., "Effect of column parameters of

FRP-confined concrete," Journal of Composites for Construction, ASCE, vol. 2, no. 4, pp.

175-185, 1998.

[26

]

Rochette, P.; Labossiere, P., "Axial testing of rectangular columns models confiend with

composites," Journal of Composites for Construction, ASCE, vol. 4, no. 3, pp. 129-136,

102

2000.

[27

]

Xiao, Y.; Wu, H., "Compressive behavior of concrete confined by carbon fiber composites

jackets," Journal of Materials in Civil Engineering, ASCE, vol. 12, no. 2, pp. 139-146, 2000.

[28

]

Wu, G.; Lu, Z.T.; Wu, Z.S., "Strength and ductility of concrete cylinders confined with FRP

composites," Journal of Construction and Building Materials, vol. 20, no. 3, pp. 134-148,

2006.

[29

]

Wei, Hua; Wu, Zhimin; Guo, Xia; Yi, Fumin, "Experimental study on partially deteriorated

strength concrete columns confined with CFRP," Engineering Structures, vol. 31, pp.

2495-2505, 2009.

[30

]

Kumutha, R.; Vaidyanathan, R.; and Palanichamy, M.S., "Behavior of reinforced concrete

rectangular columns strengthened using GFRP," Cement & Concrete Composites, vol. 29,

no. 8, pp. 609-615, Septemeber 2007.

[31

]

Hadi, M.N.S., "Comparative study of eccentrically loaded FRP wrapped columns,"

Composite Structures, vol. 74, no. 2, pp. 127-135, July 2006.

[32

]

Chaallal, Omar; shahawy, Mohsen; Hassan, Munzer, "Performance of Axially Loaded

Short Rectangular Columns Strengthened with Fiber Reinforced Polymer Wrapping,"

Journal of Composites for Construction, vol. 7, no. 3, pp. 200-208, 2003.

[33

]

Silva, Manuel A.G., "Behavior of square and circular columns strengthened with aramidic

or carbon fibers," Construction and Building Materials, vol. 25, no. 8, pp. 3222-3228,

2011.

[34

]

Toutanji, H.; and Deng, Y., "Strength and durability performance of concrete axially

loaded members confined with AFRP composites sheets," Composites Part B:

Engineering, vol. 33, no. 4, pp. 255-261, June 2002.

[35

]

Wei, H.; Wu, Z.; Guo, X.; and Yi, F., "Experimental Study in partially deteriorated strength

concrete columns confined with CFRP," Engineering Structures, vol. 21, no. 10, pp. 2495-

2505, October 2009.

[36

]

El-Hacha, R.; Green, M.F; and Wight, G. R., "Effect of Severe Environmental Exposures on

CFRP," Journal of Composites for Construction, vol. 14, no. 1, pp. 83-93,

January/February 2010.

[37

]

GRP Tubing(PTY) Ltd., [Online]. Available:

http://grptubing.com/index.php?option=com_content&view=article&id=50&Itemid=10

9. [Accessed 12 December 2011].

[38 Pessiki, S.; Harris, K. ; Kestner, J. ; Sause, R. ; and Ricles, J.M., "The Axial Behaviour of

103

] Concrete Confined with Fiber Reinforced Composites Jackets," Journal of Composites in

Construction, ASCE, vol. 5, no. 4, pp. 237-245, 2001.

[39

]

Wang, Y.C., and Restrepo, J.I., "Investigation of Concentrically Loaded Reinforced

Concrete Columns Confined with Glass Fiber-Reinforced Polymer jackets," ACI Structural

Journal, vol. 98, no. 3, pp. 377-385, 2001.

[40

]

Harries, K.A., and Carey. S.A., "Shape and Gap Effects on the Behavior of Varibly

Confined Concrete," Cement and Concrete Research, vol. 33, no. 6, pp. 881-890, 2003.

[41

]

Youssef, M.N., "Stress Strain Model for Concrete Confined by FRP Composites," PhD

dissertation, University of California-Irvine, Irvine, CA, 2003.

[42

]

Rocca. S., Galati, N., and Nanni, A., "Review of Design Guidlines for FRP Confinement of

Reinforced Concrete Columns of Noncircular Cross Sections," Journal of Composites for

Construction, ASCE, vol. 12, no. 1, pp. 80-92, 2008.

[43

]

Lam, L.; Teng, J., "Design-Oriented Stress-Strain Model for FRP-Confined Concrete in

Rectangular Columns," Jouranl of Reinforced Plastics and Composites, vol. 22, no. 13, pp.

1149-1186, 2003a.

[44

]

Lam, L., and Teng, J., "Design-Oriented Stress-Strain Model for FRP-Confined Concrete in

Rectangular Columns," Journal of Reinforced Plastics and Composites, vol. 22, no. 13, pp.

1149-1186, 2003b.

[45

]

Richart, F.E., Brandtzaeg, A., and Brown, R.L., "The Failure of Plain and Spirally

Reinforced Concrete in Compression," Bulletin No. 190, University of Illinois Engineering

Experiment Station, Urbana, USA, 1929.

[46

]

Shehata, I. A. E. M., Carneiro, L. A. V., and Shehata, L. C. D., "Strength of Short Concrete

Columns Confined With," Materials and Structures, vol. 35, pp. 50-58, 2002.

104

VITA

Haider Osamah Al-Karaghool was born on March 22nd

1987 in Baghdad, Iraq.

He started his education in Nablus elementary school in Al Yarmouk, Baghdad and

later joined Al-Mamoun secondary school, Al Mamoun, Baghdad, where he

completed Grade 7 and 8. He eventually migrated to Ajman, United Arab Emirates

where he completed Grade 9, 10 and 11 from Ajman Private School and finally Grade

12 at Al-Shoula Private School, Sharjah.

In January 2006, he pursued his bachelor degree in Civil Engineering from

American University of Sharjah, graduating in December 2009 before embarking on a

Master degree program in Civil Engineering from the same university in January

2010. During his master studies, he also worked as a Graduate Teaching Assistant

(GTA) at the university, where he assisted the lab instructor in explaining and

demonstrating experiments procedures for the construction materials lab and

surveying lab, respectively.

105

Appendix A Extra Figures and Diagrams

Figure A-1: N vs NC1

Figure A-2: N vs NC2

106

Figure A-3: N vs NG1

Figure A-4: N vs NG2

107

Figure A-5: U vs UC1

Figure A-6: U vs UC2

108

Figure A-7: U vs UG1

Figure A-8: U vs UG2

109

Figure A-9: NC1 vs UC1

Figure A-10: NC2 vs UC2

110

Figure A-11: NG1 vs UG1

Figure A-12: NG2 vs UG2

111

Figure A-13: NC1 vs NG1

Figure A-14: NC2 vs NG2

112

Figure A-15: UC1 vs UG1

Figure A-16: UC2 vs UG2

113

Figure A-17: N vs UC1

Figure A-18: N vs UC2

114

Figure A-19: N vs UG1

Figure A-20: N vs UG2

115

Figure A-21: NC1 vs NC2

Figure A-22: NG1 vs NG2

116

Figure A-23: UC1 vs UC2

Figure A-24: UG1 vs UG2

117

Appendix B Notations

N: Columns that follow normal design method

U: Column that follow under design

NC1 Normal Designed columns wrapped with one layer of CFRP

NC2 Normal Designed columns wrapped with two layer of CFRP

NG1 Normal Designed columns wrapped with one layer of GFRP

NG2 Normal Designed columns wrapped with two layer of GFRP

UC1 Under Designed columns wrapped with one layer of CFRP

UC2 Under Designed columns wrapped with two layer of CFRP

UG1 Under Designed columns wrapped with one layer of GFRP

UG2 Under Designed columns wrapped with two layer of GFRP

∆u The deflection at which the column reaches the ultimate load (mm)

∆y The deflection at which the column reaches the yield load (mm)

μ∆ Ductility index

fl Lateral confining pressure contributed by the FRP wrap

f’co Compressive strength of unconfined concrete

f’cc Compressive strength of confined concrete

Documents

STRENGTH AND DUCTILITY OF AXIALLY LOADED RC SHORT COLUMNS